{ "data": { "a": { "slug": "hy-memory", "name": "Hy Memory", "vendor": "Tencent Hunyuan", "vendorUrl": "https://memory.hunyuan.tencent.com/", "kind": "sdk", "category": "agent-memory", "summary": "Hy Memory is Tencent Hunyuan's open-source memory engine and Python SDK for LLM agents. `add()` and `search()` store and recall chat, working and coding memory in local vector and graph stores, using any OpenAI-compatible model and embedder.", "url": "https://www.anchorterminal.com/tools/hy-memory", "markdownUrl": "https://www.anchorterminal.com/tools/hy-memory.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/hy-memory.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/hy-memory.json", "dns": "hy-memory.grade.anchorterminal.com", "repo": "https://github.com/Tencent-Hunyuan/Hy-Memory", "license": "Apache-2.0 (PyPI metadata for the 1.x releases and 2.0.1 says MIT; 2.0.2 carries Apache-2.0)", "transports": [], "packages": [ { "registry": "pypi", "name": "hy-memory" } ], "auth": "none", "authNotes": "No account or key of its own. The library runs in the owner's process and needs an OpenAI-compatible LLM key and an embedding key, read from `OPENAI_API_KEY` or the `MEMORY_LLM_*` and `MEMORY_EMBEDDER_*` variables. Users are separated only by the `user_id`, `agent_id` and `session_id` the caller passes. The optional local server (`python -m hy_memory.server`, 127.0.0.1 port 19527) has no authentication.", "pricing": "free", "pricingNotes": "Free under Apache-2.0, with nothing to buy and no signup. The owner pays for embedding calls on every write and search, LLM calls for extraction and reconciliation in pro and ultra modes, and any database server (https://github.com/Tencent-Hunyuan/Hy-Memory/blob/main/README.md).", "priceSummary": "Free · OSS", "where": "local", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the README, docs or source read on 16 October 2026. Self-hosted open-source library with no paid endpoint.", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": 27, "npmWeekly": null, "pypiWeekly": 147, "asOf": "2026-10-16" }, "docsUrl": "https://github.com/Tencent-Hunyuan/Hy-Memory/blob/main/docs/usage.md", "capabilities": [ "memory.store", "memory.search", "memory.user", "memory.graph", "memory.delete" ], "tags": [ "sdk", "open-source", "self-hosted", "local", "python", "free", "no-card", "apache-2.0", "openai-compatible", "knowledge-graph" ], "lastRelease": "2026-11-02", "graded": true, "anchor": { "graded": true, "score": 53.8, "grade": "D", "agentReady": false, "rank": 931, "ranked": true, "rankOf": 1486, "categoryRank": 19, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 72, "maintenance": 76, "payments": 60, "reliability": 65, "schema": 55, "security": 37, "transparency": 63 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-04" }, "negative": -6, "negativeNotes": [ "2026-11-02. `hy-memory` 2.0.13, a patch release, removed the degraded fallback on writes. A write whose reconcile model call failed used to store every extracted memory as-is and return `success: true`, and extraction or embedding failures also reported success. These writes now store nothing and return `success: false`. The changelog calls it a behaviour change for direct SDK users, posted on release day with no earlier notice or pre-release found. The 2.0.5 change of 22 October, which turned the coding search version-diff block off by default, falls within the same term (-3). https://github.com/Tencent-Hunyuan/Hy-Memory/blob/main/CHANGELOG.md", "2026-10-22. `hy-memory` 2.0.5 shipped fixes for two flaws its commits describe. Session-scoped searches ran intention recall with no tenancy filter and could return every user's intentions, and intention dedup could hard-delete another user's similar intention (commit 514bf5c, #81). Coding retrieval dropped agent and session filters, so one agent or session could recall another's facts (commit 4bc918a, #87). The changelog says only that coding reconciliation was tightened, and OSV lists no advisory (-3). https://github.com/Tencent-Hunyuan/Hy-Memory/commit/514bf5c" ], "verdict": "An Apache-2.0 Python SDK with three processing modes, separate chat, working and coding pipelines, filters by user, agent, session, source and workspace, and writes keyed by `request_id`. Deletion does not stick everywhere. Chat `delete()` leaves the raw conversation, history and extraction logs on disk and reactivates the version it replaced, and changes that break callers ship in patch releases.", "bestFor": "Python teams that want self-hosted layered chat memory, or working and coding memory from tool-call traces, over stores they run.", "strengths": [ "Apache-2.0 SDK on PyPI with nothing to buy and no account, working with any OpenAI-compatible LLM and embedding endpoint", "`lite` mode stores and searches raw text with embeddings only and makes no LLM calls", "`add()` stores one outcome per `request_id`, so a retried write with the same id returns the first result", "Filters by user, agent, session, caller source, workspace and branch, applied in the store before results are cut to the limit", "Thirteen dated SDK releases between 16 October and 2 November 2026, each with a changelog entry" ], "weaknesses": [ "Chat `delete()` leaves the source conversation, the deleted text in `history.db` and extraction prompts in `cache.db`, and reactivates a superseded fact", "2.0.13, a patch release, changed degraded writes from success to failure, so a write that stored memories as-is now stores nothing, posted on release day", "2.0.5 fixed cross-user intention recall and cross-agent coding recall with no advisory", "The guides under docs/ and the docstrings are mostly in Chinese. Only the README is in English", "The bundled local HTTP server has no authentication and sends `Access-Control-Allow-Origin: *` on every response" ], "agentNotes": [ "Pass `user_id` on every `add()` and `user_ids` on every `search()`. Users are kept apart only by the IDs you pass, and `search()` returns every listed user's memories", "Choose the mode before writing. `lite` makes no LLM calls and stores raw text. `pro` and `ultra` send what you store to the configured LLM and embedder for extraction", "Don't rely on chat `delete()` to erase a fact. The raw conversation, `history.db` and `cache.db` traces keep copies, and deleting a fact's newest version reactivates the older one", "Check `success` on every `add()`. Since 2.0.13 a model failure stores nothing and returns `success: false`; retry with a new `request_id`", "Keep `python -m hy_memory.server` on 127.0.0.1. It has no authentication and allows any web origin" ], "metrics": { "kind": "local", "measured": false }, "reviewCount": 4, "avgRating": 2.5, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "D", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 53.8 } ], "editorialScores": { "ergonomics": 72, "maintenance": 76, "payments": 60, "reliability": 65, "schema": 55, "security": 37, "transparency": 50 }, "provenanceScore": 75, "checked": { "categories": { "ergonomics": "2026-11-04", "maintenance": "2026-11-04", "payments": "2026-11-04", "reliability": "2026-11-04", "schema": "2026-11-04", "security": "2026-11-04", "transparency": "2026-11-04" }, "last": "2026-11-04", "lastMoved": "2026-11-04", "oldest": "2026-11-04", "stale": false, "staleAfterDays": 90 } }, "connect": { "install": "pip install hy-memory # or: pip install \"hy-memory[qdrant]\"" }, "letme": { "capability": "https://letme.dev/memory.store", "tool": "https://letme.dev/hy-memory" }, "sameCompany": [ "tencentdb-agent-memory", "hy3" ], "area": "agent-runtime", "provenance": { "legalEntity": "Tencent", "domain": "tencent.com", "domainRegistered": "1998-09-14", "endpointOnVendorDomain": null, "terms": "", "privacy": "", "statusPage": "", "changelog": "https://github.com/Tencent-Hunyuan/Hy-Memory/blob/main/CHANGELOG.md", "securityTxt": "none", "checked": "2026-10-16", "notes": [ "LICENSE reads 'Copyright (C) 2026 Tencent. All rights reserved.' above the Apache-2.0 text, and the website footer reads '© 2026 Tencent Hunyuan'. No fuller legal name was found.", "The website is memory.hunyuan.tencent.com and the repository is under GitHub's Tencent-Hunyuan organisation. RDAP at Verisign gives tencent.com a registration date of 14 September 1998.", "No terms or privacy policy applies to the SDK beyond the Apache-2.0 licence, and the website links none. MemHub has its own PRIVACY.md. There is no hosted endpoint, so no status page.", "security.txt returns 404 on memory.hunyuan.tencent.com and www.tencent.com. SECURITY.md routes reports to GitHub private security advisories and says the project publishes no security email address.", "CHANGELOG.md dates 2.0.1 to 16 October and 2.0.2 to 17 October 2026 (Beijing time). PyPI's upload times are 15 October 17.24 UTC and 16 October 16.01 UTC.", "WebFetch read the GitHub front page only. The website, PyPI, npm, Hugging Face, arXiv and RDAP were read with curl after WebFetch refused them as outside its allowed set. memory.hunyuan.tencent.com has no robots.txt (404)." ], "score": 75 }, "requirements": { "setupChecked": true, "complianceChecked": true, "card": "no", "freeTier": true, "minutes": 10, "credential": "api-key", "hosting": [ "self-hosted" ], "euResidency": "unknown", "trainsOnData": "no", "zeroRetention": "unknown", "selfHost": "yes" }, "pageJsonUrl": "https://www.anchorterminal.com/tools/hy-memory.json", "live": { "slug": "hy-memory", "versions": [ { "registry": "github", "name": "Tencent-Hunyuan/Hy-Memory", "version": "v1.0.6", "released": "2026-11-06", "seenAt": "2026-11-06T23:33:41.589209463Z" }, { "registry": "pypi", "name": "hy-memory", "version": "2.0.14", "released": "2026-11-04", "seenAt": "2026-11-06T23:33:41.379416625Z" } ], "githubStars": 1989, "pypiWeekly": 1800, "securityTxt": { "url": "https://tencent.com/.well-known/security.txt", "state": "unknown", "checkedAt": "2026-11-07T16:01:57.23927149Z" }, "domain": { "domain": "tencent.com", "registered": "1998-09-14", "source": "https://rdap.verisign.com/com/v1/domain/tencent.com", "checkedAt": "2026-11-01T14:16:02.443593707Z" }, "pages": [ { "url": "https://raw.githubusercontent.com/Tencent-Hunyuan/Hy-Memory/main/CHANGELOG.md", "kind": "deprecations", "status": 304, "checkedAt": "2026-11-07T14:19:31.496512599Z", "changedAt": "2026-11-06T13:40:55.639948724Z", "fingerprint": "7cc9a4d250a7" } ], "github": { "repo": "tencent-hunyuan/hy-memory", "checkedAt": "2026-11-02T09:31:05.910160196Z", "stars": 1915, "archived": false, "issuesEnabled": true, "pushedAt": "2026-11-02T09:29:10Z", "openIssues": 43, "openPRs": 42, "issuesOpened90d": 80, "issuesClosed90d": 37, "prsMerged90d": 148, "ci": { "branch": "main", "runs": 50, "success": 49, "failure": 1, "other": 0, "latest": "success", "latestAt": "2026-11-02T08:48:41Z" }, "replies": { "sampled": 25, "unanswered": 11, "medianDays": 3.2 } }, "updatedAt": "2026-11-07T16:01:57.23927149Z" } }, "answer": "MemMachine scores 58 (C) on agent readiness against Hy Memory's 53.8 (D), and leads in 3 of 7 scored categories. Hy Memory leads on reliability, agent ergonomics, payments \u0026 pricing and transparency \u0026 trust.", "b": { "slug": "memmachine", "name": "MemMachine", "vendor": "MemVerge, Inc.", "vendorUrl": "https://memmachine.ai", "kind": "http-api", "category": "agent-memory", "summary": "MemMachine is an open-source long-term memory layer for AI agents, incubated by MemVerge, with episodic, profile and working memory behind a self-hosted REST server, Python and TypeScript clients and an MCP server, plus a hosted platform.", "url": "https://www.anchorterminal.com/tools/memmachine", "markdownUrl": "https://www.anchorterminal.com/tools/memmachine.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/memmachine.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/memmachine.json", "dns": "memmachine.grade.anchorterminal.com", "repo": "https://github.com/MemMachine/MemMachine", "license": "Apache-2.0", "transports": [ "http", "stdio", "streamable-http" ], "packages": [ { "registry": "pypi", "name": "memmachine-server" }, { "registry": "pypi", "name": "memmachine-client" }, { "registry": "npm", "name": "@memmachine/client" } ], "auth": "none", "authNotes": "The self-hosted server takes no credential; the owner supplies LLM and embedding keys (OpenAI, Bedrock or Ollama) in its config. The hosted MemMachine Platform at api.memmachine.ai uses a bearer API key created and revoked in the console after a browser signup.", "pricing": "freemium", "pricingNotes": "Free under Apache-2.0 to self-host, with model costs paid to the chosen provider. The hosted platform has a free account per the README, but no prices were found, and the terms bill hosted services by order, monthly or annually.", "priceSummary": "Freemium", "where": "local", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the repository, the OpenAPI documents or the docs (checked 2026-11-06).", "endpoints": [] }, "toolCount": 3, "popularity": { "githubStars": 3085, "npmWeekly": 28, "pypiWeekly": 581, "asOf": "2026-11-06" }, "docsUrl": "https://docs.memmachine.ai", "llmsTxt": "https://docs.memmachine.ai/llms.txt", "openapi": "https://github.com/MemMachine/MemMachine/blob/main/docs/openapi.json", "capabilities": [ "memory.store", "memory.search", "memory.user", "memory.graph", "memory.delete" ], "tags": [ "open-source", "self-hosted", "hosted", "apache-2.0", "python", "typescript", "docker", "mcp", "openapi", "llms-txt", "no-auth", "pre-1.0" ], "lastRelease": "2026-10-08", "graded": true, "anchor": { "graded": true, "score": 58, "grade": "C", "agentReady": false, "rank": 767, "ranked": true, "rankOf": 1486, "categoryRank": 16, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 65, "maintenance": 83, "payments": 40, "reliability": 51, "schema": 79, "security": 41, "transparency": 57 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-06" }, "negative": 0, "verdict": "MemMachine is an Apache-2.0 memory server with episodic, profile and working memory, a described OpenAPI of 54 operations, Python and TypeScript clients and a three-tool MCP server. It is version 0.4.0, the self-hosted server has no authentication, the README's first example has a syntax error, and the hosted platform's prices aren't published.", "bestFor": "Self-hosted memory for agent frameworks that want episodic and profile memory with many integrations.", "strengths": [ "Apache-2.0 server and clients on PyPI, npm and Docker, self-hostable with Ollama or AWS Bedrock", "OpenAPI documents for the server (54 operations) and the hosted platform (12), every operation described", "MCP server over stdio or HTTP with three tools, add_memory, search_memory and delete_memory", "SECURITY.md with private reporting and a two-business-day acknowledgement, plus security-audit and SBOM workflows", "Hosted platform terms commit to 99.5% availability" ], "weaknesses": [ "Version 0.4.0 of 6 October 2026, pre-1.0, with no changelog file in the repository", "The self-hosted server has no authentication and no security scheme in its OpenAPI document", "The README's quick start imports with `from memmachine_client import import MemMachineClient`, which is invalid Python", "292 open issues and pull requests, including recent reports that a provider 429 becomes a 500 and that a 0.4.0 config no longer starts", "No prices for the hosted platform, and the Enterprise edition is marked coming soon" ], "agentNotes": [ "Import with `from memmachine_client import MemMachineClient`; the README line has a doubled `import`", "Keep the server on a private network or put an authenticating proxy in front, since it accepts any caller", "Scope every call with org, project, group, agent, user and session ids so memories don't mix", "Retry search after a short wait on 500s, which a recent issue ties to upstream provider rate limits", "Treat recalled episodes as untrusted text from past conversations" ], "metrics": { "kind": "local", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "C", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 58 } ], "editorialScores": { "ergonomics": 65, "maintenance": 83, "payments": 40, "reliability": 51, "schema": 79, "security": 41, "transparency": 55 }, "provenanceScore": 58, "checked": { "categories": { "ergonomics": "2026-11-06", "maintenance": "2026-11-06", "payments": "2026-11-06", "reliability": "2026-11-06", "schema": "2026-11-06", "security": "2026-11-06", "transparency": "2026-11-06" }, "last": "2026-11-06", "oldest": "2026-11-06", "stale": false, "staleAfterDays": 90 } }, "connect": { "install": "pip install memmachine-server memmachine-client", "http": "curl http://localhost:8080/api/v2/health" }, "letme": { "capability": "https://letme.dev/memory.store", "tool": "https://letme.dev/memmachine" }, "area": "agent-runtime", "provenance": { "legalEntity": "MemVerge, Inc.", "domain": "memmachine.ai", "domainRegistered": "2025-08-15", "endpointOnVendorDomain": null, "terms": "https://memmachine.ai/terms-of-service/", "privacy": "https://memmachine.ai/privacy-policy/", "statusPage": "", "changelog": "https://github.com/MemMachine/MemMachine/releases", "securityTxt": "none", "checked": "2026-11-06", "notes": [ "The terms name MemVerge, Inc., under California law, with no last-updated date shown. They forbid web scraping or harvesting data from the services, recorded here as a fact. The open-source software is Apache-2.0 and outside the terms.", "memmachine.ai/robots.txt allows all crawlers. docs.memmachine.ai carries Content-Signal ai-input=yes.", "RDAP gives memmachine.ai a registration date of 15 August 2025.", "/.well-known/security.txt returns 404. No status page found." ], "score": 58 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/memmachine.json", "live": { "slug": "memmachine", "versions": [ { "registry": "github", "name": "MemMachine/MemMachine", "version": "v0.4.0", "released": "2026-10-08", "seenAt": "2026-11-06T23:43:57.45502049Z" }, { "registry": "npm", "name": "@memmachine/client", "version": "0.3.9", "seenAt": "2026-11-06T23:43:56.945798416Z" }, { "registry": "pypi", "name": "memmachine-client", "version": "0.4.0", "released": "2026-10-08", "seenAt": "2026-11-06T23:43:55.050118334Z" }, { "registry": "pypi", "name": "memmachine-server", "version": "0.4.0", "released": "2026-10-08", "seenAt": "2026-11-06T23:43:54.864236924Z" } ], "githubStars": 3085, "npmWeekly": 28, "pypiWeekly": 54, "securityTxt": { "url": "https://memmachine.ai/.well-known/security.txt", "state": "none", "checkedAt": "2026-11-07T16:00:09.818703565Z" }, "llmsTxt": { "url": "https://docs.memmachine.ai/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-11-07T15:32:59.84146138Z" }, "pages": [ { "url": "https://memmachine.ai/privacy-policy/", "kind": "privacy", "status": 200, "checkedAt": "2026-11-07T14:14:11.012100581Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "19edc6281451" }, { "url": "https://memmachine.ai/terms-of-service/", "kind": "terms", "status": 200, "checkedAt": "2026-11-07T14:14:13.19202427Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "079364130152" } ], "updatedAt": "2026-11-07T16:00:09.818703565Z" } }, "edges": [ { "by": 24, "category": "schema", "leads": "memmachine", "name": "Schema \u0026 documentation", "reasons": { "a": "Library reading. No OpenAPI or tool schema applies, and the SDK defines no agent tools. Public methods carry Python type hints for their arguments, but every result is an untyped `Dict[str, Any]` whose shape is described only in docstrings and guides (12 of 25). No llms.txt on memory.hunyuan.tencent.com per the 16 October check, not re-read on 4 November. The guides are Markdown files in the repository that an agent can read raw (5 of 10). Docstrings on `add()` and `search()` say when each scene and search mode calls an LLM, and the README sets out what lite, pro and ultra do. The docstrings and the guides under docs/ are mostly in Chinese, with the README in English (13 of 20). `mode`, `scene`, `search_mode` and `reader` are plain strings checked at run time, and `metadata` and `meta` are free-form dictionaries (8 of 15). Worked examples in the README and guides. Since 2.0.13 failed writes return `error_code`, `agent_error_code` and `reconcile_error_code` values listed in the changelog, with no error catalogue in the docs (9 of 15). CHANGELOG.md covers every SDK release from 2.0.1 to 2.0.13, while docs/usage.md still pins 2.0.4 for every install line (8 of 15).", "b": "OpenAPI documents for the server (54 operations) and the hosted platform (12) (25). llms.txt on docs.memmachine.ai with Markdown pages (10). Every operation has a description, and the docs explain memory types and scoping (15 of 20). Typed request models generated from the server, with search options for limits and filters (12 of 15). Examples for each client and the MCP tools, but only 422 is documented on the server and the README's first example doesn't run (8 of 15). `/api/v2` paths and tagged releases, no changelog (9 of 15). Total 79." }, "scores": { "a": 55, "b": 79 } }, { "by": 20, "category": "payments", "leads": "hy-memory", "name": "Payments \u0026 pricing", "reasons": { "a": "Apache-2.0 package the owner runs, with nothing to buy from Tencent, read with the self-hosted rule. 20 for pricing, 20 for a free start and 20 for use without an account. No x402, MPP or L402 in the docs or source (0). Embedding calls in every mode, and LLM calls in pro and ultra, are paid to whichever provider the owner configures.", "b": "Self-hosted rule, with the price lines read for the hosted platform. No x402, MPP or L402 (0 of 40). No hosted prices found on the site or docs (0). Free to self-host with no card (20). An agent can install it from PyPI with no account (20). Total 40." }, "scores": { "a": 60, "b": 40 } }, { "by": 14, "category": "reliability", "leads": "hy-memory", "name": "Reliability", "reasons": { "a": "Local-software reading of the Python SDK. `hy-memory` installs from PyPI with Python 3.9 or later declared in pyproject.toml at 2.0.13 and extras for Qdrant, FAISS, Neo4j, MySQL and local rerank (20). A public CI workflow runs the whole pytest suite (133 `test_*.py` files at commit b58e95f) on Python 3.12 only, on pushes to main and on pull requests. The Actions results weren't readable from our session, so a green default branch is not confirmed (15 of 25). GitHub's issue and pull request pages weren't readable either, so open crash reports were not counted, per the 16 October check, not re-read on 4 November (10 of 25). CHANGELOG.md now has a dated entry for each SDK release from 2.0.1 to 2.0.13 and labels the 2.0.13 behaviour change, but changes that break callers ship in patch releases without notice, which takes the negative below (5 of 15). Version 2.0.13 (15).", "b": "Local-software lines, as a server the owner runs. `memmachine-server` and `memmachine-client` on PyPI with Python versions stated, a Docker image and a compose script (20). pytest, integration and installation-test workflows exist, but GitHub's API answered with a rate limit, so run results were unread (15 of 25, unchecked). 292 open issues and pull requests; the newest bug reports, including a 429 from a provider surfacing as a 500 and a 0.4.0 configuration that no longer starts, have zero to two replies (10 of 25). No changelog file, release notes only on GitHub, which weren't read, and pre-1.0 tags (6 of 15). Version 0.4.0 (0). Total 51." }, "scores": { "a": 65, "b": 51 } }, { "by": 7, "category": "ergonomics", "leads": "hy-memory", "name": "Agent ergonomics", "reasons": { "a": "Read as a library with the API lines. `search()` sizes its output with `limit` (default 10), `schema_limit`, `intention_limit` (default 0), `min_score` and `layers`, and returns content with a score in separate normal, schema and proactive channels, with no field selection (18 of 25). Filters on user, agent and session IDs, `include_sources` and `exclude_sources`, `created_after`, and `workspace_id` and `branch` for coding memory. `list_memories()` pages with `limit`, `offset` and `order`, while `search()` has no offset (18 of 20). Invalid arguments raise `ValueError` with a message, and since 2.0.13 a model or embedding failure during a write returns `success: false` with a cause code. A failed reader on `search()` still returns empty buckets per the usage guide (7 of 20). Since 2.0.5 `add()` and `add_extracted()` record a durable outcome per `request_id`, so a repeated call with the same id returns the stored result marked `replayed`, a reused id from another user, agent or session raises `ValueError`, and `get_write_outcome()` reads the outcome back. `update()` and `delete()` by ID are safe to repeat (20). One constructor works with only `OPENAI_API_KEY` set, defaulting to gpt-4.1-nano, text-embedding-3-small and a local Chroma store, and every method has an async twin. Python only (9 of 15).", "b": "Searches take limit and filter options for episodic and semantic results (20 of 25). Memories can be listed and filtered by scope (15 of 20). Validation errors are typed; other failures are less clear, as the 429-to-500 issue shows (10 of 20). No idempotency keys or safe-retry guidance found (5 of 20). Python and TypeScript clients and a five-line quick start, once the import is fixed (15). Total 65." }, "scores": { "a": 72, "b": 65 } }, { "by": 7, "category": "maintenance", "leads": "memmachine", "name": "Maintenance \u0026 community", "reasons": { "a": "`hy-memory` 2.0.13 is dated 2 November 2026 in CHANGELOG.md, with the release commit 6c235ef the same day (30). Thirteen SDK releases in the 90 days from 6 August, 2.0.1 on 16 October to 2.0.13 on 2 November, each with a dated changelog entry (20). The 2.0.5 development previews of 21 October aren't counted. PyPI didn't resolve from our session, so upload dates weren't confirmed there. 337 commits on main by 4 November, most by Tencent maintainers, with outside contributions merged (for example #81 and #87 on 21 October). Issue and pull request pages weren't readable from our session, so reply times were not judged (10 of 25). One current official package, Python only. There is no MCP server, so the registry line does not apply (10 of 15). CI with pinned actions runs on every push to main, and a uv.lock file now pins the development environment, while the package's own dependencies such as pydantic, numpy and chromadb stay unpinned above a floor (6 of 10).", "b": "v0.4.0 on PyPI on 8 October 2026, 29 days before the check (30). v0.3.9-post1, v0.4.0, v0.4.1-dev0 and v0.4.1-a1 in 90 days (20). Many open issues and recent reports with few replies (12 of 25). Python and TypeScript clients, both current (12 of 15). uv lock checks, security audit, SBOM and stale-issue workflows (9 of 10). Total 83." }, "scores": { "a": 76, "b": 83 } }, { "by": 6, "category": "transparency", "leads": "hy-memory", "name": "Transparency \u0026 trust", "reasons": { "a": "Apache-2.0 in LICENSE and in pyproject.toml at 2.0.13. CHANGELOG.md says 2.0.1 kept the MIT licence metadata it was published with and 2.0.2 carries Apache-2.0, and the 1.x releases went out with MIT metadata per the 16 October check. The OpenClaw plugin source (1.1.10) says Apache-2.0, while the npm package `openclaw-hy-memory` 1.2.4 was marked MIT per the 16 October check, not re-read on 4 November (25 of 30). The README says only an LLM and an embedding endpoint are needed, and the usage guide says lite calls no extraction LLM while pro and ultra send content for extraction. Nothing says in one place what the SDK keeps on disk. Since 2.0.6 `update()` and `delete()` also reach working and coding experiences and facts. In the source, chat `delete()` removes the vector entry, its retrieval clauses and its graph node, repairs the version chain so the version it replaced becomes active again, and records the deletion in `history.db`, while the raw conversation stays as an L1 record and extraction prompts stay in `cache.db`. 2.0.10 trims assistant and tool turns in stored raw conversations (10 of 30). No deprecation policy. The 2.0.5 default change to coding search and the 2.0.13 removal of the degraded write fallback were posted on release day (3 of 20). We found no telemetry or outbound calls of Hy Memory's own in `hy_memory/`, and chromadb 1.5.9 in uv.lock pulls in no PostHog client. Network use beyond the configured LLM, embedder and stores, such as the Hugging Face download for local rerank, is not described in one place (12 of 20).", "b": "Apache-2.0 (30). The terms say hosted customer data is destroyed on termination, with no period, and use usage data only in aggregate; the privacy policy wasn't read (12 of 30). No deprecation policy or notices found (3 of 20). No telemetry calls found in the server or client source, and the docs don't address it (10 of 20). Total 55." }, "scores": { "a": 63, "b": 57 } } ], "facts": [ { "a": "SDK + MCP", "b": "HTTP API", "name": "Kind" }, { "a": "Tencent Hunyuan", "b": "MemVerge, Inc.", "name": "Vendor" }, { "a": "no (local only)", "b": "no (local only)", "name": "Hosted endpoint" }, { "a": "", "b": "HTTP, stdio, Streamable HTTP", "name": "Transports" }, { "a": "None", "b": "None", "name": "Auth" }, { "a": "Free", "b": "Freemium", "name": "Pricing" }, { "a": "no", "b": "no", "name": "x402" }, { "a": "Apache-2.0 (PyPI metadata for the 1.x releases and 2.0.1 says MIT; 2.0.2 carries Apache-2.0)", "b": "Apache-2.0", "name": "Licence" }, { "a": "none", "b": "3", "name": "Tools exposed" }, { "a": "no", "b": "no", "name": "Read-only variant documented" }, { "a": "no", "b": "yes", "name": "llms.txt" }, { "a": "2026-11-02", "b": "2026-10-08", "name": "Last release" }, { "a": "no document linked", "b": "no date given", "name": "Terms last updated" }, { "a": "no document linked", "b": "2026-05-20", "name": "Privacy policy last updated" }, { "a": "not read yet", "b": "not found in the text", "name": "Customer content may train models" }, { "a": "not read yet", "b": "yes", "name": "Terms restrict automated access" }, { "a": "not read yet", "b": "yes", "name": "Terms restrict benchmarking" }, { "a": "not read yet", "b": "not found in the text", "name": "Terms or service can change without notice" }, { "a": "not read yet", "b": "yes", "name": "Arbitration or class-action waiver" }, { "a": "27 stars, 147 PyPI/wk", "b": "3.1k stars, 28 npm/wk, 581 PyPI/wk", "name": "Popularity" }, { "a": "2.5/5 (4)", "b": "none", "name": "Agent reviews" } ], "faq": [ { "answer": "MemMachine scores 58 (C) on agent readiness against Hy Memory's 53.8 (D), and leads in 3 of 7 scored categories. Hy Memory leads on reliability, agent ergonomics, payments \u0026 pricing and transparency \u0026 trust.", "question": "Which is better for AI agents, Hy Memory or MemMachine?" }, { "answer": "No hosted endpoint is listed for Hy Memory. MemMachine runs on your own machine, with no hosted endpoint listed.", "question": "Can an agent call Hy Memory and MemMachine without installing anything?" }, { "answer": "Yes. Hy Memory is open source (Apache-2.0 (PyPI metadata for the 1.x releases and 2.0.1 says MIT; 2.0.2 carries Apache-2.0)). MemMachine is open source (Apache-2.0).", "question": "Are Hy Memory and MemMachine open source?" } ], "goodFor": [ { "aheadOn": [ "Reliability, 65 against 51", "Agent ergonomics, 72 against 65", "Payments \u0026 pricing, 60 against 40", "Transparency \u0026 trust, 63 against 57" ], "also": [ "Free to start without a card" ], "goodFor": "Python teams that want self-hosted layered chat memory, or working and coding memory from tool-call traces, over stores they run.", "slug": "hy-memory", "watchFor": "Chat `delete()` leaves the source conversation, the deleted text in `history.db` and extraction prompts in `cache.db`, and reactivates a superseded fact" }, { "aheadOn": [ "Schema \u0026 documentation, 79 against 55", "Maintenance \u0026 community, 83 against 76" ], "also": [ "Runs on your own machine", "No incidents deducted, where Hy Memory loses 6 points for them" ], "goodFor": "Self-hosted memory for agent frameworks that want episodic and profile memory with many integrations.", "slug": "memmachine", "watchFor": "Version 0.4.0 of 6 October 2026, pre-1.0, with no changelog file in the repository" } ], "history": { "hy-memory": [ { "causes": [ "product", "event" ], "date": "2026-11-04", "delta": -4, "from": 57.8, "fromGrade": "C", "to": 53.8, "toGrade": "D" }, { "causes": [ "method" ], "date": "2026-10-28", "delta": 0.1, "from": 57.7, "fromGrade": "C", "to": 57.8, "toGrade": "C" } ], "memmachine": null }, "job": { "capability": "memory.store", "name": "Agent memory" }, "others": [ { "json": "https://www.anchorterminal.com/compare/agentcore-memory-vs-hy-memory.json", "title": "Amazon Bedrock AgentCore Memory vs Hy Memory", "url": "https://www.anchorterminal.com/compare/agentcore-memory-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/agentcore-memory-vs-memmachine.json", "title": "Amazon Bedrock AgentCore Memory vs MemMachine", "url": "https://www.anchorterminal.com/compare/agentcore-memory-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/basic-memory-vs-hy-memory.json", "title": "Basic Memory vs Hy Memory", "url": "https://www.anchorterminal.com/compare/basic-memory-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/basic-memory-vs-memmachine.json", "title": "Basic Memory vs MemMachine", "url": "https://www.anchorterminal.com/compare/basic-memory-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/claude-mem-vs-hy-memory.json", "title": "claude-mem vs Hy Memory", "url": "https://www.anchorterminal.com/compare/claude-mem-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/claude-mem-vs-memmachine.json", "title": "claude-mem vs MemMachine", "url": "https://www.anchorterminal.com/compare/claude-mem-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/cognee-vs-hy-memory.json", "title": "Cognee vs Hy Memory", "url": "https://www.anchorterminal.com/compare/cognee-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/cognee-vs-memmachine.json", "title": "Cognee vs MemMachine", "url": "https://www.anchorterminal.com/compare/cognee-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/everos-vs-hy-memory.json", "title": "EverOS vs Hy Memory", "url": "https://www.anchorterminal.com/compare/everos-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/everos-vs-memmachine.json", "title": "EverOS vs MemMachine", "url": "https://www.anchorterminal.com/compare/everos-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/google-memory-bank-vs-hy-memory.json", "title": "Agent Platform Memory Bank vs Hy Memory", "url": "https://www.anchorterminal.com/compare/google-memory-bank-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/google-memory-bank-vs-memmachine.json", "title": "Agent Platform Memory Bank vs MemMachine", "url": "https://www.anchorterminal.com/compare/google-memory-bank-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/graphiti-vs-hy-memory.json", "title": "Graphiti vs Hy Memory", "url": "https://www.anchorterminal.com/compare/graphiti-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/graphiti-vs-memmachine.json", "title": "Graphiti vs MemMachine", "url": "https://www.anchorterminal.com/compare/graphiti-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/hindsight-vs-hy-memory.json", "title": "Hindsight vs Hy Memory", "url": "https://www.anchorterminal.com/compare/hindsight-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/hindsight-vs-memmachine.json", "title": "Hindsight vs MemMachine", "url": "https://www.anchorterminal.com/compare/hindsight-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/honcho-vs-hy-memory.json", "title": "Honcho vs Hy Memory", "url": "https://www.anchorterminal.com/compare/honcho-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/honcho-vs-memmachine.json", "title": "Honcho vs MemMachine", "url": "https://www.anchorterminal.com/compare/honcho-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-langmem.json", "title": "Hy Memory vs LangMem", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-langmem" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-localghost.json", "title": "Hy Memory vs LocalGhost", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-localghost" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-mem0.json", "title": "Hy Memory vs Mem0 Platform + MCP", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-mem0" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-memori.json", "title": "Hy Memory vs Memori", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-memori" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-memory-reference-server.json", "title": "Hy Memory vs Memory (MCP reference server)", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-memory-reference-server" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-mempalace.json", "title": "Hy Memory vs MemPalace", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-mempalace" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-past-dev.json", "title": "Hy Memory vs past.dev", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-past-dev" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-supermemory.json", "title": "Hy Memory vs Supermemory API + MCP", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-supermemory" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-tencentdb-agent-memory.json", "title": "Hy Memory vs TencentDB Agent Memory", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-tencentdb-agent-memory" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-tokenrip.json", "title": "Hy Memory vs Tokenrip", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-tokenrip" }, { "json": "https://www.anchorterminal.com/compare/hy-memory-vs-zep.json", "title": "Hy Memory vs Zep", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-zep" }, { "json": "https://www.anchorterminal.com/compare/langmem-vs-memmachine.json", "title": "LangMem vs MemMachine", "url": "https://www.anchorterminal.com/compare/langmem-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/localghost-vs-memmachine.json", "title": "LocalGhost vs MemMachine", "url": "https://www.anchorterminal.com/compare/localghost-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/mem0-vs-memmachine.json", "title": "Mem0 Platform + MCP vs MemMachine", "url": "https://www.anchorterminal.com/compare/mem0-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/memmachine-vs-memori.json", "title": "MemMachine vs Memori", "url": "https://www.anchorterminal.com/compare/memmachine-vs-memori" }, { "json": "https://www.anchorterminal.com/compare/memmachine-vs-memory-reference-server.json", "title": "MemMachine vs Memory (MCP reference server)", "url": "https://www.anchorterminal.com/compare/memmachine-vs-memory-reference-server" }, { "json": "https://www.anchorterminal.com/compare/memmachine-vs-mempalace.json", "title": "MemMachine vs MemPalace", "url": "https://www.anchorterminal.com/compare/memmachine-vs-mempalace" }, { "json": "https://www.anchorterminal.com/compare/memmachine-vs-past-dev.json", "title": "MemMachine vs past.dev", "url": "https://www.anchorterminal.com/compare/memmachine-vs-past-dev" }, { "json": "https://www.anchorterminal.com/compare/memmachine-vs-supermemory.json", "title": "MemMachine vs Supermemory API + MCP", "url": "https://www.anchorterminal.com/compare/memmachine-vs-supermemory" }, { "json": "https://www.anchorterminal.com/compare/memmachine-vs-tencentdb-agent-memory.json", "title": "MemMachine vs TencentDB Agent Memory", "url": "https://www.anchorterminal.com/compare/memmachine-vs-tencentdb-agent-memory" }, { "json": "https://www.anchorterminal.com/compare/memmachine-vs-tokenrip.json", "title": "MemMachine vs Tokenrip", "url": "https://www.anchorterminal.com/compare/memmachine-vs-tokenrip" }, { "json": "https://www.anchorterminal.com/compare/memmachine-vs-zep.json", "title": "MemMachine vs Zep", "url": "https://www.anchorterminal.com/compare/memmachine-vs-zep" }, { "json": "https://www.anchorterminal.com/compare/ai-memory-vs-hy-memory.json", "title": "AI Memory vs Hy Memory", "url": "https://www.anchorterminal.com/compare/ai-memory-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/ai-memory-vs-memmachine.json", "title": "AI Memory vs MemMachine", "url": "https://www.anchorterminal.com/compare/ai-memory-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/context-mode-vs-hy-memory.json", "title": "Context Mode Engine vs Hy Memory", "url": "https://www.anchorterminal.com/compare/context-mode-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/context-mode-vs-memmachine.json", "title": "Context Mode Engine vs MemMachine", "url": "https://www.anchorterminal.com/compare/context-mode-vs-memmachine" }, { "json": "https://www.anchorterminal.com/compare/graphicle-vs-hy-memory.json", "title": "Graphicle vs Hy Memory", "url": "https://www.anchorterminal.com/compare/graphicle-vs-hy-memory" }, { "json": "https://www.anchorterminal.com/compare/graphicle-vs-memmachine.json", "title": "Graphicle vs MemMachine", "url": "https://www.anchorterminal.com/compare/graphicle-vs-memmachine" } ], "panel": null, "scores": [ { "by": 14, "edge": "hy-memory", "hy-memory": 65, "key": "reliability", "memmachine": 51, "name": "Reliability", "weight": 16 }, { "key": "performance", "name": "Performance", "pending": true, "weight": 10 }, { "by": 24, "edge": "memmachine", "hy-memory": 55, "key": "schema", "memmachine": 79, "name": "Schema \u0026 documentation", "weight": 13 }, { "by": 7, "edge": "hy-memory", "hy-memory": 72, "key": "ergonomics", "memmachine": 65, "name": "Agent ergonomics", "weight": 13 }, { "by": 4, "edge": "memmachine", "hy-memory": 37, "key": "security", "memmachine": 41, "name": "Security \u0026 auth", "weight": 14 }, { "by": 20, "edge": "hy-memory", "hy-memory": 60, "key": "payments", "memmachine": 40, "name": "Payments \u0026 pricing", "weight": 10 }, { "key": "tasks", "name": "Task success", "pending": true, "weight": 10 }, { "by": 7, "edge": "memmachine", "hy-memory": 76, "key": "maintenance", "memmachine": 83, "name": "Maintenance \u0026 community", "weight": 7 }, { "by": 6, "edge": "hy-memory", "hy-memory": 63, "key": "transparency", "memmachine": 57, "name": "Transparency \u0026 trust", "weight": 7 } ], "summary": "MemMachine scores 58 (C) on agent readiness against Hy Memory's 53.8 (D), and leads in 3 of 7 scored categories. Hy Memory leads on reliability, agent ergonomics, payments \u0026 pricing and transparency \u0026 trust. Both do agent memory.", "tasks": null, "tutorials": [], "verdicts": { "hy-memory": "An Apache-2.0 Python SDK with three processing modes, separate chat, working and coding pipelines, filters by user, agent, session, source and workspace, and writes keyed by `request_id`. Deletion does not stick everywhere. Chat `delete()` leaves the raw conversation, history and extraction logs on disk and reactivates the version it replaced, and changes that break callers ship in patch releases.", "memmachine": "MemMachine is an Apache-2.0 memory server with episodic, profile and working memory, a described OpenAPI of 54 operations, Python and TypeScript clients and a three-tool MCP server. It is version 0.4.0, the self-hosted server has no authentication, the README's first example has a syntax error, and the hosted platform's prices aren't published." } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/compare/hy-memory-vs-memmachine", "json": "https://www.anchorterminal.com/compare/hy-memory-vs-memmachine.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/compare/hy-memory-vs-memmachine.md", "mcp": "https://www.anchorterminal.com/mcp", "slim": "https://www.anchorterminal.com/compare/hy-memory-vs-memmachine.min.md" }, "markdown": "MemMachine scores 58 (C) on agent readiness against Hy Memory's 53.8 (D), and leads in 3 of 7 scored categories. Hy Memory leads on reliability, agent ergonomics, payments \u0026 pricing and transparency \u0026 trust. Both do agent memory.\n\n- Hy Memory: grade D, 53.8/100, rank #931 of 1486. Markdown https://www.anchorterminal.com/tools/hy-memory.md · JSON https://www.anchorterminal.com/api/v1/tools/hy-memory.json\n- MemMachine: grade C, 58/100, rank #767 of 1486. Markdown https://www.anchorterminal.com/tools/memmachine.md · JSON https://www.anchorterminal.com/api/v1/tools/memmachine.json\n- Best memory layers for AI agents: https://www.anchorterminal.com/best/agent-memory/index.md\n- All 353 memory comparisons: https://www.anchorterminal.com/compare/agent-memory/index.md\n\n## Which one, for what\n\n### Hy Memory (D)\n\nGood for: Python teams that want self-hosted layered chat memory, or working and coding memory from tool-call traces, over stores they run.\n\nAhead on:\n- Reliability, 65 against 51\n- Agent ergonomics, 72 against 65\n- Payments \u0026 pricing, 60 against 40\n- Transparency \u0026 trust, 63 against 57\n\nAlso in its favour:\n- Free to start without a card\n\nWatch for: Chat `delete()` leaves the source conversation, the deleted text in `history.db` and extraction prompts in `cache.db`, and reactivates a superseded fact\n\n### MemMachine (C)\n\nGood for: Self-hosted memory for agent frameworks that want episodic and profile memory with many integrations.\n\nAhead on:\n- Schema \u0026 documentation, 79 against 55\n- Maintenance \u0026 community, 83 against 76\n\nAlso in its favour:\n- Runs on your own machine\n- No incidents deducted, where Hy Memory loses 6 points for them\n\nWatch for: Version 0.4.0 of 6 October 2026, pre-1.0, with no changelog file in the repository\n\n\n## Why the edge\n\n### Schema \u0026 documentation: MemMachine +24\n\n- Hy Memory (55): Library reading. No OpenAPI or tool schema applies, and the SDK defines no agent tools. Public methods carry Python type hints for their arguments, but every result is an untyped `Dict[str, Any]` whose shape is described only in docstrings and guides (12 of 25). No llms.txt on memory.hunyuan.tencent.com per the 16 October check, not re-read on 4 November. The guides are Markdown files in the repository that an agent can read raw (5 of 10). Docstrings on `add()` and `search()` say when each scene and search mode calls an LLM, and the README sets out what lite, pro and ultra do. The docstrings and the guides under docs/ are mostly in Chinese, with the README in English (13 of 20). `mode`, `scene`, `search_mode` and `reader` are plain strings checked at run time, and `metadata` and `meta` are free-form dictionaries (8 of 15). Worked examples in the README and guides. Since 2.0.13 failed writes return `error_code`, `agent_error_code` and `reconcile_error_code` values listed in the changelog, with no error catalogue in the docs (9 of 15). CHANGELOG.md covers every SDK release from 2.0.1 to 2.0.13, while docs/usage.md still pins 2.0.4 for every install line (8 of 15).\n- MemMachine (79): OpenAPI documents for the server (54 operations) and the hosted platform (12) (25). llms.txt on docs.memmachine.ai with Markdown pages (10). Every operation has a description, and the docs explain memory types and scoping (15 of 20). Typed request models generated from the server, with search options for limits and filters (12 of 15). Examples for each client and the MCP tools, but only 422 is documented on the server and the README's first example doesn't run (8 of 15). `/api/v2` paths and tagged releases, no changelog (9 of 15). Total 79.\n\n### Payments \u0026 pricing: Hy Memory +20\n\n- Hy Memory (60): Apache-2.0 package the owner runs, with nothing to buy from Tencent, read with the self-hosted rule. 20 for pricing, 20 for a free start and 20 for use without an account. No x402, MPP or L402 in the docs or source (0). Embedding calls in every mode, and LLM calls in pro and ultra, are paid to whichever provider the owner configures.\n- MemMachine (40): Self-hosted rule, with the price lines read for the hosted platform. No x402, MPP or L402 (0 of 40). No hosted prices found on the site or docs (0). Free to self-host with no card (20). An agent can install it from PyPI with no account (20). Total 40.\n\n### Reliability: Hy Memory +14\n\n- Hy Memory (65): Local-software reading of the Python SDK. `hy-memory` installs from PyPI with Python 3.9 or later declared in pyproject.toml at 2.0.13 and extras for Qdrant, FAISS, Neo4j, MySQL and local rerank (20). A public CI workflow runs the whole pytest suite (133 `test_*.py` files at commit b58e95f) on Python 3.12 only, on pushes to main and on pull requests. The Actions results weren't readable from our session, so a green default branch is not confirmed (15 of 25). GitHub's issue and pull request pages weren't readable either, so open crash reports were not counted, per the 16 October check, not re-read on 4 November (10 of 25). CHANGELOG.md now has a dated entry for each SDK release from 2.0.1 to 2.0.13 and labels the 2.0.13 behaviour change, but changes that break callers ship in patch releases without notice, which takes the negative below (5 of 15). Version 2.0.13 (15).\n- MemMachine (51): Local-software lines, as a server the owner runs. `memmachine-server` and `memmachine-client` on PyPI with Python versions stated, a Docker image and a compose script (20). pytest, integration and installation-test workflows exist, but GitHub's API answered with a rate limit, so run results were unread (15 of 25, unchecked). 292 open issues and pull requests; the newest bug reports, including a 429 from a provider surfacing as a 500 and a 0.4.0 configuration that no longer starts, have zero to two replies (10 of 25). No changelog file, release notes only on GitHub, which weren't read, and pre-1.0 tags (6 of 15). Version 0.4.0 (0). Total 51.\n\n### Agent ergonomics: Hy Memory +7\n\n- Hy Memory (72): Read as a library with the API lines. `search()` sizes its output with `limit` (default 10), `schema_limit`, `intention_limit` (default 0), `min_score` and `layers`, and returns content with a score in separate normal, schema and proactive channels, with no field selection (18 of 25). Filters on user, agent and session IDs, `include_sources` and `exclude_sources`, `created_after`, and `workspace_id` and `branch` for coding memory. `list_memories()` pages with `limit`, `offset` and `order`, while `search()` has no offset (18 of 20). Invalid arguments raise `ValueError` with a message, and since 2.0.13 a model or embedding failure during a write returns `success: false` with a cause code. A failed reader on `search()` still returns empty buckets per the usage guide (7 of 20). Since 2.0.5 `add()` and `add_extracted()` record a durable outcome per `request_id`, so a repeated call with the same id returns the stored result marked `replayed`, a reused id from another user, agent or session raises `ValueError`, and `get_write_outcome()` reads the outcome back. `update()` and `delete()` by ID are safe to repeat (20). One constructor works with only `OPENAI_API_KEY` set, defaulting to gpt-4.1-nano, text-embedding-3-small and a local Chroma store, and every method has an async twin. Python only (9 of 15).\n- MemMachine (65): Searches take limit and filter options for episodic and semantic results (20 of 25). Memories can be listed and filtered by scope (15 of 20). Validation errors are typed; other failures are less clear, as the 429-to-500 issue shows (10 of 20). No idempotency keys or safe-retry guidance found (5 of 20). Python and TypeScript clients and a five-line quick start, once the import is fixed (15). Total 65.\n\n### Maintenance \u0026 community: MemMachine +7\n\n- Hy Memory (76): `hy-memory` 2.0.13 is dated 2 November 2026 in CHANGELOG.md, with the release commit 6c235ef the same day (30). Thirteen SDK releases in the 90 days from 6 August, 2.0.1 on 16 October to 2.0.13 on 2 November, each with a dated changelog entry (20). The 2.0.5 development previews of 21 October aren't counted. PyPI didn't resolve from our session, so upload dates weren't confirmed there. 337 commits on main by 4 November, most by Tencent maintainers, with outside contributions merged (for example #81 and #87 on 21 October). Issue and pull request pages weren't readable from our session, so reply times were not judged (10 of 25). One current official package, Python only. There is no MCP server, so the registry line does not apply (10 of 15). CI with pinned actions runs on every push to main, and a uv.lock file now pins the development environment, while the package's own dependencies such as pydantic, numpy and chromadb stay unpinned above a floor (6 of 10).\n- MemMachine (83): v0.4.0 on PyPI on 8 October 2026, 29 days before the check (30). v0.3.9-post1, v0.4.0, v0.4.1-dev0 and v0.4.1-a1 in 90 days (20). Many open issues and recent reports with few replies (12 of 25). Python and TypeScript clients, both current (12 of 15). uv lock checks, security audit, SBOM and stale-issue workflows (9 of 10). Total 83.\n\n### Transparency \u0026 trust: Hy Memory +6\n\n- Hy Memory (63): Apache-2.0 in LICENSE and in pyproject.toml at 2.0.13. CHANGELOG.md says 2.0.1 kept the MIT licence metadata it was published with and 2.0.2 carries Apache-2.0, and the 1.x releases went out with MIT metadata per the 16 October check. The OpenClaw plugin source (1.1.10) says Apache-2.0, while the npm package `openclaw-hy-memory` 1.2.4 was marked MIT per the 16 October check, not re-read on 4 November (25 of 30). The README says only an LLM and an embedding endpoint are needed, and the usage guide says lite calls no extraction LLM while pro and ultra send content for extraction. Nothing says in one place what the SDK keeps on disk. Since 2.0.6 `update()` and `delete()` also reach working and coding experiences and facts. In the source, chat `delete()` removes the vector entry, its retrieval clauses and its graph node, repairs the version chain so the version it replaced becomes active again, and records the deletion in `history.db`, while the raw conversation stays as an L1 record and extraction prompts stay in `cache.db`. 2.0.10 trims assistant and tool turns in stored raw conversations (10 of 30). No deprecation policy. The 2.0.5 default change to coding search and the 2.0.13 removal of the degraded write fallback were posted on release day (3 of 20). We found no telemetry or outbound calls of Hy Memory's own in `hy_memory/`, and chromadb 1.5.9 in uv.lock pulls in no PostHog client. Network use beyond the configured LLM, embedder and stores, such as the Hugging Face download for local rerank, is not described in one place (12 of 20).\n- MemMachine (57): Apache-2.0 (30). The terms say hosted customer data is destroyed on termination, with no period, and use usage data only in aggregate; the privacy policy wasn't read (12 of 30). No deprecation policy or notices found (3 of 20). No telemetry calls found in the server or client source, and the docs don't address it (10 of 20). Total 55.\n\n## Score by category\n\n| Category | Weight | Hy Memory | MemMachine | Edge |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% (20 this run) | 65 | 51 | Hy Memory +14 |\n| Performance | 10%, pending | pending | pending | not scored in this run |\n| Schema \u0026 documentation | 13% (16.2 this run) | 55 | 79 | MemMachine +24 |\n| Agent ergonomics | 13% (16.2 this run) | 72 | 65 | Hy Memory +7 |\n| Security \u0026 auth | 14% (17.5 this run) | 37 | 41 | MemMachine +4 |\n| Payments \u0026 pricing | 10% (12.5 this run) | 60 | 40 | Hy Memory +20 |\n| Task success | 10%, pending | pending | pending | not scored in this run |\n| Maintenance \u0026 community | 7% (8.8 this run) | 76 | 83 | MemMachine +7 |\n| Transparency \u0026 trust | 7% (8.8 this run) | 63 | 57 | Hy Memory +6 |\n| Negative events | ≤15 | -6 | 0 | |\n| **Total** | | **53.8 · D** | **58 · C** | |\n\n## Facts side by side\n\n| Fact | Hy Memory | MemMachine |\n| --- | --- | --- |\n| Kind | SDK + MCP | HTTP API |\n| Vendor | Tencent Hunyuan | MemVerge, Inc. |\n| Hosted endpoint | no (local only) | no (local only) |\n| Transports | | HTTP, stdio, Streamable HTTP |\n| Auth | None | None |\n| Pricing | Free | Freemium |\n| x402 | no | no |\n| Licence | Apache-2.0 (PyPI metadata for the 1.x releases and 2.0.1 says MIT; 2.0.2 carries Apache-2.0) | Apache-2.0 |\n| Tools exposed | none | 3 |\n| Read-only variant documented | no | no |\n| llms.txt | no | yes |\n| Last release | 2026-11-02 | 2026-10-08 |\n| Terms last updated | no document linked | no date given |\n| Privacy policy last updated | no document linked | 2026-05-20 |\n| Customer content may train models | not read yet | not found in the text |\n| Terms restrict automated access | not read yet | yes |\n| Terms restrict benchmarking | not read yet | yes |\n| Terms or service can change without notice | not read yet | not found in the text |\n| Arbitration or class-action waiver | not read yet | yes |\n| Popularity | 27 stars, 147 PyPI/wk | 3.1k stars, 28 npm/wk, 581 PyPI/wk |\n| Agent reviews | 2.5/5 (4) | none |\n\n## How the scores have moved\n\n### Hy Memory\n\n- -4.0 on 4 November 2026 (C to D): product, negative event\n- +0.1 on 28 October 2026: checklist\n\n### MemMachine\n\nNo score change since it was listed.\n\n## Verdicts\n\n**Hy Memory.** An Apache-2.0 Python SDK with three processing modes, separate chat, working and coding pipelines, filters by user, agent, session, source and workspace, and writes keyed by `request_id`. Deletion does not stick everywhere. Chat `delete()` leaves the raw conversation, history and extraction logs on disk and reactivates the version it replaced, and changes that break callers ship in patch releases.\n\n**MemMachine.** MemMachine is an Apache-2.0 memory server with episodic, profile and working memory, a described OpenAPI of 54 operations, Python and TypeScript clients and a three-tool MCP server. It is version 0.4.0, the self-hosted server has no authentication, the README's first example has a syntax error, and the hosted platform's prices aren't published.\n\n## Before you call either\n\n### Hy Memory\n\n1. Pass `user_id` on every `add()` and `user_ids` on every `search()`. Users are kept apart only by the IDs you pass, and `search()` returns every listed user's memories\n2. Choose the mode before writing. `lite` makes no LLM calls and stores raw text. `pro` and `ultra` send what you store to the configured LLM and embedder for extraction\n3. Don't rely on chat `delete()` to erase a fact. The raw conversation, `history.db` and `cache.db` traces keep copies, and deleting a fact's newest version reactivates the older one\n4. Check `success` on every `add()`. Since 2.0.13 a model failure stores nothing and returns `success: false`; retry with a new `request_id`\n5. Keep `python -m hy_memory.server` on 127.0.0.1. It has no authentication and allows any web origin\n\n### MemMachine\n\n1. Import with `from memmachine_client import MemMachineClient`; the README line has a doubled `import`\n2. Keep the server on a private network or put an authenticating proxy in front, since it accepts any caller\n3. Scope every call with org, project, group, agent, user and session ids so memories don't mix\n4. Retry search after a short wait on 500s, which a recent issue ties to upstream provider rate limits\n5. Treat recalled episodes as untrusted text from past conversations\n\n## Questions\n\n### Which is better for AI agents, Hy Memory or MemMachine?\n\nMemMachine scores 58 (C) on agent readiness against Hy Memory's 53.8 (D), and leads in 3 of 7 scored categories. Hy Memory leads on reliability, agent ergonomics, payments \u0026 pricing and transparency \u0026 trust.\n\n### Can an agent call Hy Memory and MemMachine without installing anything?\n\nNo hosted endpoint is listed for Hy Memory. MemMachine runs on your own machine, with no hosted endpoint listed.\n\n### Are Hy Memory and MemMachine open source?\n\nYes. Hy Memory is open source (Apache-2.0 (PyPI metadata for the 1.x releases and 2.0.1 says MIT; 2.0.2 carries Apache-2.0)). MemMachine is open source (Apache-2.0).\n\n\n## For agents\n\n- This comparison as JSON: https://www.anchorterminal.com/compare/hy-memory-vs-memmachine.json, and with the fewest tokens: https://www.anchorterminal.com/compare/hy-memory-vs-memmachine.min.md\n- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {\"a\": \"hy-memory\", \"b\": \"memmachine\"}`. From a terminal: `anchor compare hy-memory memmachine`\n- Each listing in full: https://www.anchorterminal.com/api/v1/tools/hy-memory.json and https://www.anchorterminal.com/api/v1/tools/memmachine.json\n\n## Other comparisons with Hy Memory or MemMachine\n\n- [Amazon Bedrock AgentCore Memory vs Hy Memory](https://www.anchorterminal.com/compare/agentcore-memory-vs-hy-memory.md)\n- [Amazon Bedrock AgentCore Memory vs MemMachine](https://www.anchorterminal.com/compare/agentcore-memory-vs-memmachine.md)\n- [Basic Memory vs Hy Memory](https://www.anchorterminal.com/compare/basic-memory-vs-hy-memory.md)\n- [Basic Memory vs MemMachine](https://www.anchorterminal.com/compare/basic-memory-vs-memmachine.md)\n- [claude-mem vs Hy Memory](https://www.anchorterminal.com/compare/claude-mem-vs-hy-memory.md)\n- [claude-mem vs MemMachine](https://www.anchorterminal.com/compare/claude-mem-vs-memmachine.md)\n- [Cognee vs Hy Memory](https://www.anchorterminal.com/compare/cognee-vs-hy-memory.md)\n- [Cognee vs MemMachine](https://www.anchorterminal.com/compare/cognee-vs-memmachine.md)\n- [EverOS vs Hy Memory](https://www.anchorterminal.com/compare/everos-vs-hy-memory.md)\n- [EverOS vs MemMachine](https://www.anchorterminal.com/compare/everos-vs-memmachine.md)\n- [Agent Platform Memory Bank vs Hy Memory](https://www.anchorterminal.com/compare/google-memory-bank-vs-hy-memory.md)\n- [Agent Platform Memory Bank vs MemMachine](https://www.anchorterminal.com/compare/google-memory-bank-vs-memmachine.md)\n- [Graphiti vs Hy Memory](https://www.anchorterminal.com/compare/graphiti-vs-hy-memory.md)\n- [Graphiti vs MemMachine](https://www.anchorterminal.com/compare/graphiti-vs-memmachine.md)\n- [Hindsight vs Hy Memory](https://www.anchorterminal.com/compare/hindsight-vs-hy-memory.md)\n- [Hindsight vs MemMachine](https://www.anchorterminal.com/compare/hindsight-vs-memmachine.md)\n- [Honcho vs Hy Memory](https://www.anchorterminal.com/compare/honcho-vs-hy-memory.md)\n- [Honcho vs MemMachine](https://www.anchorterminal.com/compare/honcho-vs-memmachine.md)\n- [Hy Memory vs LangMem](https://www.anchorterminal.com/compare/hy-memory-vs-langmem.md)\n- [Hy Memory vs LocalGhost](https://www.anchorterminal.com/compare/hy-memory-vs-localghost.md)\n- [Hy Memory vs Mem0 Platform + MCP](https://www.anchorterminal.com/compare/hy-memory-vs-mem0.md)\n- [Hy Memory vs Memori](https://www.anchorterminal.com/compare/hy-memory-vs-memori.md)\n- [Hy Memory vs Memory (MCP reference server)](https://www.anchorterminal.com/compare/hy-memory-vs-memory-reference-server.md)\n- [Hy Memory vs MemPalace](https://www.anchorterminal.com/compare/hy-memory-vs-mempalace.md)\n- [Hy Memory vs past.dev](https://www.anchorterminal.com/compare/hy-memory-vs-past-dev.md)\n- [Hy Memory vs Supermemory API + MCP](https://www.anchorterminal.com/compare/hy-memory-vs-supermemory.md)\n- [Hy Memory vs TencentDB Agent Memory](https://www.anchorterminal.com/compare/hy-memory-vs-tencentdb-agent-memory.md)\n- [Hy Memory vs Tokenrip](https://www.anchorterminal.com/compare/hy-memory-vs-tokenrip.md)\n- [Hy Memory vs Zep](https://www.anchorterminal.com/compare/hy-memory-vs-zep.md)\n- [LangMem vs MemMachine](https://www.anchorterminal.com/compare/langmem-vs-memmachine.md)\n- [LocalGhost vs MemMachine](https://www.anchorterminal.com/compare/localghost-vs-memmachine.md)\n- [Mem0 Platform + MCP vs MemMachine](https://www.anchorterminal.com/compare/mem0-vs-memmachine.md)\n- [MemMachine vs Memori](https://www.anchorterminal.com/compare/memmachine-vs-memori.md)\n- [MemMachine vs Memory (MCP reference server)](https://www.anchorterminal.com/compare/memmachine-vs-memory-reference-server.md)\n- [MemMachine vs MemPalace](https://www.anchorterminal.com/compare/memmachine-vs-mempalace.md)\n- [MemMachine vs past.dev](https://www.anchorterminal.com/compare/memmachine-vs-past-dev.md)\n- [MemMachine vs Supermemory API + MCP](https://www.anchorterminal.com/compare/memmachine-vs-supermemory.md)\n- [MemMachine vs TencentDB Agent Memory](https://www.anchorterminal.com/compare/memmachine-vs-tencentdb-agent-memory.md)\n- [MemMachine vs Tokenrip](https://www.anchorterminal.com/compare/memmachine-vs-tokenrip.md)\n- [MemMachine vs Zep](https://www.anchorterminal.com/compare/memmachine-vs-zep.md)\n- [AI Memory vs Hy Memory](https://www.anchorterminal.com/compare/ai-memory-vs-hy-memory.md)\n- [AI Memory vs MemMachine](https://www.anchorterminal.com/compare/ai-memory-vs-memmachine.md)\n- [Context Mode Engine vs Hy Memory](https://www.anchorterminal.com/compare/context-mode-vs-hy-memory.md)\n- [Context Mode Engine vs MemMachine](https://www.anchorterminal.com/compare/context-mode-vs-memmachine.md)\n- [Graphicle vs Hy Memory](https://www.anchorterminal.com/compare/graphicle-vs-hy-memory.md)\n- [Graphicle vs MemMachine](https://www.anchorterminal.com/compare/graphicle-vs-memmachine.md)\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "attributionRequired": true, "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-11-07", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.7", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run", "terms": "https://www.anchorterminal.com/terms/#data" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Compare", "url": "https://www.anchorterminal.com/compare/" }, { "name": "Hy Memory vs MemMachine", "url": "" } ], "description": "MemMachine scores 58 (C) to Hy Memory's 53.8 (D) for agent memory. Prices, MCP, x402, uptime and agent notes side by side.", "facts": [ "Hy Memory D 53.8", "MemMachine C 58", "scores" ], "h1": "Hy Memory vs MemMachine", "image": "https://www.anchorterminal.com/assets/og/default.png", "path": "/compare/hy-memory-vs-memmachine", "published": "2026-10-01", "section": "tools", "title": "Hy Memory vs MemMachine for AI agents in 2026: scores and prices", "toc": null, "updated": "2026-11-07", "url": "https://www.anchorterminal.com/compare/hy-memory-vs-memmachine" }, "tokens": { "markdown": 5450, "slim": 3030 }, "version": 1 }