CueLight vs Inworld Realtime API

CueLight 24

Read as local software. Installs with pip from the Git repository, not from PyPI, with Python 3.10 or later and its dependencies declared in pyproject.toml, a partial held from 15 October (10 of 20). The repository holds two tests (streaming matches whole-sequence output, chunk size is enforced), but it has no GitHub workflow at all, so no public CI runs them (0). No issues or pull requests were open on 15 October, and GitHub's pages were not re-read on 4 November, so the earlier partial is held (12 of 25). Version 0.1.0 in pyproject.toml with no tags or changelog (2 of 15). Below 1.0, and the README calls it a first iteration (0). 10 + 0 + 12 + 2 + 0 = 24.

Inworld Realtime API 62

Hosted reading of the checklist. status.inworld.ai is a public status page (20), with history we couldn't read, as on inworld-tts (5). Concurrent sessions are published per plan, 10 on On-Demand up to 1,000 on Growth (15). The rate limits page gives exponential backoff with jitter and Retry-After on longer windows, and sessions emit rate_limits.updated, with nothing on reconnecting a dropped session (12 of 15). An SLA appears only in the Enterprise column, with no terms (0). No preview or beta label on the Realtime API (10). 20 + 5 + 15 + 12 + 0 + 10 = 62.

CueLight 41

Published on 14 October 2026, 21 days before this check (30). A second commit on 15 October added an 8-bit Nano ONNX model to the browser demo, with no version change, tag or changelog entry, so it isn't a release. One publication in 90 days, below the three the line needs (0). No issues or pull requests were open on 15 October, and GitHub's pages were not re-read, so responsiveness is held (5 of 25). Not on PyPI as of 15 October (pypi.org did not resolve on 4 November). The weights are on Hugging Face and the package installs only from Git (3 of 15). Dependencies are current minimums (torch 2.1 or later), with no lockfile, Dependabot or CI (3 of 10). 30 + 0 + 5 + 3 + 3 = 41.

Inworld Realtime API 71

Release note of 16 October 2026 (cache_transcript), 21 days ago (30). Three dated Realtime entries in the 90 days from 8 August, on 14 September, 23 September and 16 October (20). Discord and support@inworld.ai, with no evidence of answers read (8 of 15). Works with OpenAI's Realtime clients and the CLI is current at 1.5.0, but there is no Inworld Realtime SDK (8 of 15). CLI on npm under MIT, CI not read (5 of 10). Total 71.

CueLight 60

No payment protocol (0). Read with the self-hosted rule. Apache-2.0 code and weights with no hosted or paid option (20). No card or account (20). Weights download anonymously, with no signup (20). 0 + 20 + 20 + 20 = 60.

Inworld Realtime API 40

No x402, MPP or L402 (0). No per-session price, but each billed part (STT per hour, TTS per character, LLM per token) is priced publicly (20). On-Demand includes free STT and TTS allowances and lists no payment method, though some premium LLMs need a card on file (20). A person signs up in a browser before a key exists (0). Total 40.

CueLight 50

Library reading. No OpenAPI applies. The public classes are small, with type hints on most signatures and frozen dataclasses for both policies, though push takes an untyped chunk (15 of 25). No llms.txt on the Pages site, and docs are the README (0). The README says which model suits which device, when to choose each preset, and states the limits (English only, audio only, Base's CPU cost) (16 of 20). The 320-sample frame is enforced by an assert, while an unknown preset name falls back silently to default thresholds rather than raising (8 of 15). Quick-start snippets and two example scripts. Error behaviour is not documented (8 of 15). A version number in pyproject.toml and __version__, with no changelog (3 of 15). 15 + 0 + 16 + 8 + 8 + 3 = 50.

Inworld Realtime API 60

No AsyncAPI or other machine-readable file for the socket, and the spec paths redirect (0). llms.txt, Markdown twins and a docs MCP server (10). The overview, OpenAI migration guide, providerData reference and the billing page's account of how a session is counted say what to use when (16 of 20). The reference lists every client and server event with field types and required marks, as text rather than a schema (12 of 15). JavaScript and Python example projects and an error event with type, code and param, without a list of codes (9 of 15). Dated release notes with affected clients and required actions, but no API version beyond /v1 (13 of 15). Total 60.

CueLight 50

Local-software reading. No credential and no open port. The code runs in the caller's process, weights load from safetensors rather than pickle, but snapshot_download fetches the main branch of each Hugging Face repository with no pinned revision (20 of 30). It only classifies audio and takes no actions (16 of 20). It returns probabilities, not content from third parties (10). No logging. Per-frame probabilities and a reset counter are visible to the caller (4 of 15). No SECURITY.md, disclosure policy, security.txt or advisories (0 of 20). 20 + 16 + 10 + 4 + 0 = 50. Re-read from the source on 4 November, unchanged.

Inworld Realtime API 58

Workspace API keys sent as Basic credentials, deleted to revoke, with separate keys per environment advised. The CLI mints read-only keys unless --write is given, but write toggles cover only the Voice and Router APIs, so no key is limited to one product. Browsers use short-lived tokens minted server side. Session IDs, not secrets, go in the URL (25 of 30). Tools run in the client, so the developer approves side effects, with no approval step in the API (10 of 20). The agent listens to callers and passes their speech to an LLM, and no prompt-injection guidance was found (5 of 15). Usage per key and a Performance page with errors and throttling (8 of 15). SOC 2 Type II and HIPAA claimed and a vulnerability report form, with no security.txt or bug bounty, per the 4 November reading (10 of 20). Total 58. Zero data retention covers session content from the Developer plan up, but not requests routed to third-party LLM providers.

CueLight 61

Library reading. Four public names and one call per frame, with an 11 KB README, so the context cost of learning it is low (22 of 25). Output is one probability and one speech probability per frame, and thresholds, minimum silence, timeout and reset behaviour can all be set (14 of 20). The only error is an assert on chunk size, which python -O strips, and a mistyped preset gives defaults without a warning (6 of 20). push is stateful by design. clear() resets between calls and the detector resets itself at each new turn, and the source shows streaming matches whole-clip output to 1e-4 (10 of 20). One required argument with working defaults. Python only, with ONNX step files and a browser demo but no packaged second-language SDK (9 of 15). 22 + 14 + 6 + 10 + 9 = 61.

Inworld Realtime API 68

Scored as an API. Conversation history defaults to the last 100 items, can be trimmed, and can be cached on Anthropic models, and every response.done carries a usage block (18 of 25). No lists to page. Output controls are modalities, turn detection settings, optional transcription and response.speak (12 of 20). Error events follow OpenAI's type and code shape, unsupported fields come back as non-fatal unsupported_field warnings, and item_not_found names the bad parameter (14 of 20). Tools run on the client and a follow-up response starts automatically after a tool result. Nothing covers side effects of a tool call cut off by an interruption (12 of 20). Works with OpenAI Realtime clients after changing the URL and key, with no Inworld Realtime SDK (12 of 15). Total 68.