{ "data": { "a": { "slug": "amazon-bedrock-customization", "name": "Amazon Bedrock model customisation", "vendor": "Amazon Web Services", "vendorUrl": "https://aws.amazon.com/bedrock/", "kind": "http-api", "category": "fine-tuning", "summary": "Managed supervised fine-tuning, reinforcement fine-tuning and distillation of Amazon Nova, Meta Llama and selected open-weight models on Amazon Bedrock, run as asynchronous jobs through the Bedrock control-plane API, the AWS SDKs and CLI, or OpenAI-compatible endpoints.", "url": "https://www.anchorterminal.com/tools/amazon-bedrock-customization", "markdownUrl": "https://www.anchorterminal.com/tools/amazon-bedrock-customization.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/amazon-bedrock-customization.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/amazon-bedrock-customization.json", "transports": [ "http" ], "remoteUrl": "https://bedrock.{region}.amazonaws.com/model-customization-jobs", "packages": [ { "registry": "pypi", "name": "boto3" }, { "registry": "npm", "name": "@aws-sdk/client-bedrock" } ], "auth": "mixed", "authNotes": "AWS Signature Version 4 with IAM credentials or a role allowed to call `bedrock:CreateModelCustomizationJob`, plus `iam:PassRole` for a service role that Bedrock assumes to read training data from S3 and write output. Access is self-serve inside an AWS account. The OpenAI-compatible fine-tuning endpoints on `bedrock-mantle` also accept a Bedrock API key as a bearer token, short-term (up to 12 hours) or long-term, and reinforcement jobs need `lambda:InvokeFunction` on the reward function.", "pricing": "usage", "pricingNotes": "Supervised fine-tuning is billed per training token (dataset tokens times epochs). Nova Micro $1, Nova Lite $2, Nova 2 Lite $3.78 and Nova Pro $8 per 1M tokens, and Llama from $0.50 (3.2 1B) to $7.99 (3.1 70B). Reinforcement fine-tuning is $80 per training hour. Each custom model costs $1.95 a month to store. Inference is at base-model token prices where on-demand deployment is supported, otherwise Provisioned Throughput by the hour. No free tier for customisation appears on the pricing page. New AWS accounts get up to $200 in Free Tier credits, and there is no sandbox (https://aws.amazon.com/bedrock/pricing/).", "priceSummary": "Pay per use", "where": "hosted", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the user guide, the API reference or the pricing page (checked 2026-10-08).", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": null, "npmWeekly": 2810365, "pypiWeekly": 573748207, "asOf": "2026-10-08" }, "docsUrl": "https://docs.aws.amazon.com/bedrock/latest/userguide/custom-models.html", "llmsTxt": "https://docs.aws.amazon.com/bedrock/latest/userguide/llms.txt", "capabilities": [ "finetune.sft", "finetune.rl" ], "tags": [ "hosted", "usage-priced", "closed-source", "enterprise", "python", "typescript", "async-jobs", "llms-txt", "status-page", "soc2" ], "lastRelease": "2026-05-28", "graded": true, "anchor": { "graded": true, "score": 75.8, "grade": "BB", "agentReady": true, "rank": 39, "ranked": true, "rankOf": 842, "categoryRank": 1, "methodology": "0.4", "run": "2026-10-01", "scores": { "ergonomics": 77, "maintenance": 45, "payments": 30, "reliability": 95, "schema": 87, "security": 87, "transparency": 83 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-10-08" }, "negative": 0, "verdict": "Job creation takes an idempotency token, job lists filter and paginate, and the Service Terms give the customer exclusive use of a tuned model. Jobs run in two US Regions only, weights can't be exported, and the newest customisation API change found dates from 28 May 2026.", "bestFor": "Teams already on AWS that want to tune Amazon Nova or Llama models, or run reinforcement fine-tuning with a Lambda reward function, and serve the result inside Bedrock.", "strengths": [ "`CreateModelCustomizationJob` accepts a `clientRequestToken`, so a repeated create after a timeout doesn't start a second job", "`ListModelCustomizationJobs` filters by status, name and creation time, sorts, and pages with `maxResults` up to 1,000 and `nextToken`", "Service Terms section 50.12.4 gives the customer exclusive use of a customised model and bars third-party model providers from accessing it", "Training prices are public per 1M tokens, from $0.50 for Llama 3.2 1B to $8 for Nova Pro, with storage at $1.95 a model a month", "Supervised tunes of Nova and Llama 3.3 70B deploy for on-demand, per-token inference at base-model prices, without Provisioned Throughput" ], "weaknesses": [ "Fine-tuning runs in us-east-1 and us-west-2 only, and each base model in one of them (two for the Titan models)", "No weight export was found in the reviewed documentation, and Service Terms section 50.11 forbids extracting model weights", "Setup needs an IAM service role, `iam:PassRole` and S3 buckets for input and output before the first job", "Reinforcement fine-tuning is billed at $80 a training hour and covers three models. Distillation is not available for Anthropic models", "The Bedrock SLA is dated 4 October 2023 and doesn't mention customisation jobs. No customisation launch appears in the latest 100 Bedrock What's New posts" ], "agentNotes": [ "Send a `clientRequestToken` with every `CreateModelCustomizationJob` call, and poll `GetModelCustomizationJob`. Jobs are asynchronous and can take hours", "Create the job in the Region that hosts the base model (Nova in us-east-1, Llama and Claude 3 Haiku in us-west-2), with the S3 bucket in the same Region", "Pass an IAM service role that trusts `bedrock.amazonaws.com` and can read the training data and write the output location. The caller needs `iam:PassRole`", "After a job completes, call `CreateCustomModelDeployment` and use the deployment ARN as `modelId`. Models outside the on-demand list need Provisioned Throughput", "For gpt-oss-20b and Qwen3 32B, use `/v1/fine_tuning/jobs` on `bedrock-mantle.us-west-2.api.aws` with a Bedrock API key and a Lambda grader ARN" ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "BB", "methodology": "0.4", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 75.8 } ], "editorialScores": { "ergonomics": 77, "maintenance": 45, "payments": 30, "reliability": 95, "schema": 87, "security": 87, "transparency": 77 }, "provenanceScore": 88 }, "connect": { "install": "pip install boto3 # or: npm i @aws-sdk/client-bedrock", "http": "curl -X POST \"https://bedrock.us-east-1.amazonaws.com/model-customization-jobs\" \\\n --aws-sigv4 \"aws:amz:us-east-1:bedrock\" --user \"$AWS_ACCESS_KEY_ID:$AWS_SECRET_ACCESS_KEY\" \\\n -H \"content-type: application/json\" \\\n -d '{\"jobName\":\"my-tune\",\"customModelName\":\"my-nova-tune\",\"roleArn\":\"arn:aws:iam::123456789012:role/BedrockCustomisationRole\",\"baseModelIdentifier\":\"amazon.nova-2-lite-v1:0:256k\",\"clientRequestToken\":\"my-tune-001\",\"trainingDataConfig\":{\"s3Uri\":\"s3://my-bucket/train.jsonl\"},\"outputDataConfig\":{\"s3Uri\":\"s3://my-bucket/output/\"}}'" }, "letme": { "capability": "https://letme.dev/finetune.sft", "tool": "https://letme.dev/amazon-bedrock-customization" }, "sameCompany": [ "agentcore-code-interpreter", "amazon-textract", "aws-end-user-messaging", "amazon-sns", "amazon-s3" ], "area": "models", "unitPrices": [ { "item": "Nova 2 Lite, supervised fine-tuning", "unit": "1m-tokens", "usd": 3.78, "note": "Training tokens, counted as dataset tokens times epochs" }, { "item": "Nova Pro, supervised fine-tuning", "unit": "1m-tokens", "usd": 8, "note": "Training tokens, counted as dataset tokens times epochs" }, { "item": "Nova Lite, supervised fine-tuning", "unit": "1m-tokens", "usd": 2, "note": "Training tokens, counted as dataset tokens times epochs" }, { "item": "Nova Micro, supervised fine-tuning", "unit": "1m-tokens", "usd": 1, "note": "Training tokens, counted as dataset tokens times epochs" }, { "item": "Llama 3.1 70B Instruct, fine-tuning", "unit": "1m-tokens", "usd": 7.99, "note": "Training tokens, counted as dataset tokens times epochs" }, { "item": "Llama 3.1 8B Instruct, fine-tuning", "unit": "1m-tokens", "usd": 1.49, "note": "Training tokens, counted as dataset tokens times epochs" }, { "item": "Llama 3.2 90B Instruct, fine-tuning", "unit": "1m-tokens", "usd": 7.9, "note": "Training tokens, counted as dataset tokens times epochs" }, { "item": "Llama 3.2 11B Instruct, fine-tuning", "unit": "1m-tokens", "usd": 3.5, "note": "Training tokens, counted as dataset tokens times epochs" }, { "item": "Llama 3.2 3B Instruct, fine-tuning", "unit": "1m-tokens", "usd": 1.1, "note": "Training tokens, counted as dataset tokens times epochs" }, { "item": "Llama 3.2 1B Instruct, fine-tuning", "unit": "1m-tokens", "usd": 0.5, "note": "Training tokens, counted as dataset tokens times epochs" } ], "provenance": { "legalEntity": "Amazon Web Services, Inc.", "domain": "amazonaws.com", "domainRegistered": "2005-08-18", "domainNote": "The control-plane endpoints are `bedrock.\u003cregion\u003e.amazonaws.com`. The OpenAI-compatible fine-tuning endpoints are on `bedrock-mantle.\u003cregion\u003e.api.aws`, a second AWS domain. The security.txt on aws.amazon.com passed its Expires date on 2026-09-24.", "endpointOnVendorDomain": true, "terms": "https://aws.amazon.com/service-terms/", "privacy": "https://aws.amazon.com/privacy/", "statusPage": "https://health.aws.amazon.com/health/status", "changelog": "https://docs.aws.amazon.com/bedrock/latest/userguide/doc-history.html", "securityTxt": "expired", "checked": "2026-10-08", "notes": [ "The AWS Service Terms (last updated 1 October 2026) hold the Bedrock clauses in section 50.12, including 50.12.4 on customised models. They sit under the AWS Customer Agreement at aws.amazon.com/agreement (last updated 14 August 2026), where the contracting party depends on the account country.", "The AWS Privacy Notice was last updated on 18 May 2026 and names Amazon Web Services, Inc., 410 Terry Avenue North, Seattle.", "RDAP for amazonaws.com gives a registration date of 2005-08-18, read on 8 October 2026.", "aws.amazon.com/.well-known/security.txt carries Expires 2026-09-24T16:25:03Z and was still expired on 8 October 2026. It points to vdp.aws.security and a HackerOne disclosure programme.", "The pricing page's customisation tables are filled by script. Token prices were read from the price file the page loads from b0.p.awsstatic.com." ], "score": 88 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/amazon-bedrock-customization.json", "live": { "slug": "amazon-bedrock-customization", "probe": { "target": "https://bedrock.{region}.amazonaws.com/model-customization-jobs", "method": "get", "lastAt": "2026-10-09T11:46:21.208433901Z", "lastOk": false, "lastStatus": 0, "lastMs": 0, "lastNote": "invalid character \"{\" in host name", "authRequired": false, "uptime24h": 0, "uptime30d": 0, "p50ms24h": 0, "p95ms24h": 0, "samples24h": 44, "samples30d": 44, "days": [ { "date": "2026-10-09", "probes": 44, "ok": 0 } ], "outages": [ { "start": "2026-10-09T07:40:18.733946397Z", "end": "0001-01-01T00:00:00Z", "note": "invalid character \"{\" in host name" } ] }, "updatedAt": "2026-10-09T11:46:21.208433901Z" } }, "answer": "Amazon Bedrock model customisation scores 75.8 (BB) on agent readiness against Tinker's 51 (D), and leads in 6 of 7 scored categories. Tinker leads on maintenance \u0026 community.", "b": { "slug": "tinker", "name": "Tinker", "vendor": "Thinking Machines Lab", "vendorUrl": "https://thinkingmachines.ai/tinker/", "kind": "sdk", "category": "fine-tuning", "summary": "Thinking Machines Lab's API for model training.", "url": "https://www.anchorterminal.com/tools/tinker", "markdownUrl": "https://www.anchorterminal.com/tools/tinker.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/tinker.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/tinker.json", "repo": "https://github.com/thinking-machines-lab/tinker-cookbook", "license": "Apache-2.0 (cookbook)", "transports": [ "http" ], "packages": [ { "registry": "pypi", "name": "tinker" }, { "registry": "pypi", "name": "tinker-cookbook" } ], "auth": "api-key", "authNotes": "API key from the Tinker console, exported as `TINKER_API_KEY`, or `tinker auth login`. Sign-up is at auth.thinkingmachines.ai and the quickstart says to add payment details in Billing before training.", "pricing": "usage", "pricingNotes": "Per 1M tokens, split into prefill, cached prefill (20 per cent of prefill), sample and train. Qwen3.8-27B $1.86 prefill, $5.595 sample, $4.103 train; Qwen3.5-9B $0.66, $1.995, $1.463; GPT-OSS-20B $0.18, $0.45, $0.396; DeepSeek-V3.1 $1.695, $4.215, $3.718; Inkling $1.87, $4.68, $5.61; Inkling-Small $0.58, $1.44, $1.73. MoE models are priced by active parameters. Checkpoint storage $0.10 per GB-month. Prices rose on 2026-07-17 for standard-context models. No free credits are mentioned (https://tinker-docs.thinkingmachines.ai/tinker/models/).", "priceSummary": "Pay per use", "where": "local", "x402": { "level": "no", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": 4000, "npmWeekly": null, "pypiWeekly": 330895, "asOf": "2026-09-30" }, "docsUrl": "https://tinker-docs.thinkingmachines.ai", "llmsTxt": "https://tinker-docs.thinkingmachines.ai/llms.txt", "capabilities": [ "finetune.sft", "finetune.preference", "finetune.rl", "finetune.lora", "finetune.export" ], "tags": [ "hosted", "usage-priced", "card-required", "open-weights", "llms-txt", "python", "open-source" ], "lastRelease": "2026-09-30", "graded": true, "anchor": { "graded": true, "score": 51, "grade": "D", "agentReady": false, "rank": 677, "ranked": true, "rankOf": 842, "categoryRank": 8, "methodology": "0.4", "run": "2026-10-01", "scores": { "ergonomics": 53, "maintenance": 87, "payments": 20, "reliability": 35, "schema": 70, "security": 55, "transparency": 49 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-10-01" }, "negative": 0, "verdict": "Full control of the training loop with the GPUs abstracted away, plus recipes for SFT, DPO, RL and distillation. LoRA only; no full-parameter training.", "bestFor": "Researchers and teams writing custom post-training loops, especially RL, who want per-token billing and the weights at the end.", "strengths": [ "Full control of the training loop with the GPUs abstracted away, plus recipes for SFT, DPO, RL and distillation", "Checkpoints download and merge into Hugging Face safetensors, so the weights can leave", "Per-token billing with machine-readable prices in models.json", "Ten SDK releases in September 2026 and a dated changelog that names removals", "Audit log through the SDK for admins, and SDK retries with stable request IDs" ], "weaknesses": [ "LoRA only; no full-parameter training", "Python SDK only, with no REST reference or OpenAPI", "No terms of service, status page or SLA found", "The privacy notice (August 2025) doesn't cover training data or weights", "Standard-context prices rose on 2026-07-17, and there's no free tier" ], "agentNotes": [ "Set `TINKER_API_KEY` and start from the cookbook recipes rather than the raw primitives", "Read the 'Avoid Client-Side Timeouts and Retries' guide before wrapping sampling calls in your own retries; the SDK already retries sampling with stable request IDs", "Save intermediate checkpoints with a TTL between 1 hour and 10 years; storage bills at $0.10 a GB-month until they expire", "Read models.json for current prices before a run; sampling tokens cost more than training tokens on the open models", "Check the model deprecations page before pinning a base model; 18 were retired on 2026-06-12" ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 2, "avgRating": 3.5, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "D", "methodology": "0.4", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 51 } ], "editorialScores": { "ergonomics": 53, "maintenance": 87, "payments": 20, "reliability": 35, "schema": 70, "security": 55, "transparency": 30 }, "provenanceScore": 67 }, "connect": { "install": "uv pip install tinker tinker-cookbook # then export TINKER_API_KEY=..." }, "letme": { "capability": "https://letme.dev/finetune.sft", "tool": "https://letme.dev/tinker" }, "area": "models", "unitPrices": [ { "item": "Qwen3.8-27B, training", "unit": "1m-tokens", "usd": 4.103 }, { "item": "Qwen3.8-27B, sampling", "unit": "1m-tokens", "usd": 5.595 }, { "item": "Qwen3.8-27B, prefill", "unit": "1m-tokens", "usd": 1.86, "note": "Cached prefill $0.372" }, { "item": "Qwen3.5-9B, training", "unit": "1m-tokens", "usd": 1.463 }, { "item": "GPT-OSS-20B, training", "unit": "1m-tokens", "usd": 0.396 }, { "item": "DeepSeek-V3.1, training", "unit": "1m-tokens", "usd": 3.718 }, { "item": "Inkling, training", "unit": "1m-tokens", "usd": 5.61 }, { "item": "Inkling-Small, training", "unit": "1m-tokens", "usd": 1.73 }, { "item": "Checkpoint storage", "unit": "gb-month", "usd": 0.1 } ], "provenance": { "legalEntity": "Thinking Machines Labs, Inc.", "domain": "thinkingmachines.ai", "domainRegistered": "", "endpointOnVendorDomain": null, "terms": "", "privacy": "https://thinkingmachines.ai/privacy/", "statusPage": "", "changelog": "https://tinker-docs.thinkingmachines.ai/changelog/", "securityTxt": "valid", "checked": "2026-09-30", "notes": [ "The privacy notice (2025-08-18) names Thinking Machines Labs, Inc. as data controller and gives no address. We found no terms of service page on thinkingmachines.ai or the docs; the support page links only to email, Discord and GitHub.", "The service is reached through the SDK's ServiceClient with an undocumented default base URL, so there's no endpoint to check against the domain.", "security.txt at thinkingmachines.ai lists security-reports@thinkingmachines.ai and expires 2029-07-13.", "No status page was found.", "The .ai registry's RDAP server refused our requests, so the registration date is blank." ], "score": 67 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/tinker.json", "live": { "slug": "tinker", "versions": [ { "registry": "github", "name": "thinking-machines-lab/tinker-cookbook", "version": "v0.5.7", "released": "2026-09-03", "seenAt": "2026-10-08T16:32:11.324997657Z" }, { "registry": "pypi", "name": "tinker", "version": "0.32.0", "released": "2026-10-02", "seenAt": "2026-10-08T16:32:09.236417949Z" }, { "registry": "pypi", "name": "tinker-cookbook", "version": "0.5.7", "released": "2026-09-03", "seenAt": "2026-10-08T16:32:09.427353039Z" } ], "githubStars": 4179, "pypiWeekly": 421815, "securityTxt": { "url": "https://thinkingmachines.ai/.well-known/security.txt", "state": "valid", "expires": "2029-07-13T07:00:00.000Z", "checkedAt": "2026-10-08T15:38:33.720455026Z" }, "llmsTxt": { "url": "https://tinker-docs.thinkingmachines.ai/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-10-08T14:00:56.670858456Z" }, "domain": { "domain": "thinkingmachines.ai", "registered": "2024-07-09", "source": "https://rdap.identitydigital.services/rdap/domain/thinkingmachines.ai", "checkedAt": "2026-10-04T13:07:26.919974206Z" }, "pages": [ { "url": "https://tinker-docs.thinkingmachines.ai/changelog/", "kind": "changelog", "status": 200, "checkedAt": "2026-10-08T18:25:16.191875788Z", "changedAt": "2026-10-06T16:13:08.925985263Z", "fingerprint": "18093297ee7c" }, { "url": "https://thinkingmachines.ai/privacy/", "kind": "privacy", "status": 404, "checkedAt": "2026-10-08T18:25:14.907567846Z", "changedAt": "0001-01-01T00:00:00Z" } ], "updatedAt": "2026-10-08T18:25:16.191875788Z" } }, "facts": [ { "a": "HTTP API", "b": "SDK + MCP", "name": "Kind" }, { "a": "Amazon Web Services", "b": "Thinking Machines Lab", "name": "Vendor" }, { "a": "https://bedrock.{region}.amazonaws.com/model-customization-jobs", "b": "no (local only)", "name": "Hosted endpoint" }, { "a": "HTTP", "b": "HTTP", "name": "Transports" }, { "a": "OAuth or key", "b": "API key", "name": "Auth" }, { "a": "Pay per use", "b": "Pay per use", "name": "Pricing" }, { "a": "no", "b": "no", "name": "x402" }, { "a": "none", "b": "Apache-2.0 (cookbook)", "name": "Licence" }, { "a": "no", "b": "no", "name": "Read-only variant documented" }, { "a": "yes", "b": "yes", "name": "llms.txt" }, { "a": "2026-05-28", "b": "2026-09-30", "name": "Last release" }, { "a": "2026-10-01", "b": "no document linked", "name": "Terms last updated" }, { "a": "2026-05-18", "b": "couldn't be read", "name": "Privacy policy last updated" }, { "a": "yes, with an opt-out", "b": "", "name": "Customer content may train models" }, { "a": "yes", "b": "", "name": "Terms restrict automated access" }, { "a": "yes", "b": "", "name": "Terms restrict benchmarking" }, { "a": "yes", "b": "", "name": "Terms or service can change without notice" }, { "a": "not found in the text", "b": "", "name": "Arbitration or class-action waiver" }, { "a": "2.8M npm/wk, 573.7M PyPI/wk", "b": "4k stars, 331k PyPI/wk", "name": "Popularity" }, { "a": "none", "b": "3.5/5 (2)", "name": "Agent reviews" } ], "faq": [ { "answer": "Amazon Bedrock model customisation scores 75.8 (BB) on agent readiness against Tinker's 51 (D), and leads in 6 of 7 scored categories. Tinker leads on maintenance \u0026 community.", "question": "Which is better for AI agents, Amazon Bedrock model customisation or Tinker?" }, { "answer": "Amazon Bedrock model customisation has a hosted endpoint at https://bedrock.{region}.amazonaws.com/model-customization-jobs. No hosted endpoint is listed for Tinker.", "question": "Can an agent call Amazon Bedrock model customisation and Tinker without installing anything?" }, { "answer": "No open-source release is listed for Amazon Bedrock model customisation. Tinker is open source (Apache-2.0 (cookbook)).", "question": "Are Amazon Bedrock model customisation and Tinker open source?" } ], "goodFor": [ { "aheadOn": [ "Reliability, 95 against 35", "Schema \u0026 documentation, 87 against 70", "Agent ergonomics, 77 against 53", "Security \u0026 auth, 87 against 55", "Payments \u0026 pricing, 30 against 20", "Transparency \u0026 trust, 83 against 49" ], "also": [ "Agent-ready, a grade of BB or better", "A hosted endpoint, with nothing to install" ], "goodFor": "Teams already on AWS that want to tune Amazon Nova or Llama models, or run reinforcement fine-tuning with a Lambda reward function, and serve the result inside Bedrock.", "slug": "amazon-bedrock-customization", "watchFor": "Fine-tuning runs in us-east-1 and us-west-2 only, and each base model in one of them (two for the Titan models)" }, { "aheadOn": [ "Maintenance \u0026 community, 87 against 45" ], "also": [ "Open source" ], "goodFor": "Researchers and teams writing custom post-training loops, especially RL, who want per-token billing and the weights at the end.", "slug": "tinker", "watchFor": "LoRA only; no full-parameter training" } ], "job": { "capability": "finetune.sft", "name": "Finetune sft" }, "others": [ { "json": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-axolotl.json", "title": "Amazon Bedrock model customisation vs Axolotl", "url": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-axolotl" }, { "json": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-azure-foundry-fine-tuning.json", "title": "Amazon Bedrock model customisation vs Microsoft Foundry fine-tuning (Azure OpenAI)", "url": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-azure-foundry-fine-tuning" }, { "json": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-fireworks-fine-tuning.json", "title": "Amazon Bedrock model customisation vs Fireworks AI Fine-tuning", "url": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-fireworks-fine-tuning" }, { "json": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-nebius-token-factory-fine-tuning.json", "title": "Amazon Bedrock model customisation vs Nebius Token Factory fine-tuning", "url": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-nebius-token-factory-fine-tuning" }, { "json": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-together-fine-tuning.json", "title": "Amazon Bedrock model customisation vs Together AI Fine-tuning", "url": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-together-fine-tuning" }, { "json": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-unsloth.json", "title": "Amazon Bedrock model customisation vs Unsloth", "url": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-unsloth" }, { "json": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-vertex-ai-tuning.json", "title": "Amazon Bedrock model customisation vs Vertex AI Gemini tuning", "url": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-vertex-ai-tuning" }, { "json": "https://www.anchorterminal.com/compare/axolotl-vs-tinker.json", "title": "Axolotl vs Tinker", "url": "https://www.anchorterminal.com/compare/axolotl-vs-tinker" }, { "json": "https://www.anchorterminal.com/compare/azure-foundry-fine-tuning-vs-tinker.json", "title": "Microsoft Foundry fine-tuning (Azure OpenAI) vs Tinker", "url": "https://www.anchorterminal.com/compare/azure-foundry-fine-tuning-vs-tinker" }, { "json": "https://www.anchorterminal.com/compare/fireworks-fine-tuning-vs-tinker.json", "title": "Fireworks AI Fine-tuning vs Tinker", "url": "https://www.anchorterminal.com/compare/fireworks-fine-tuning-vs-tinker" }, { "json": "https://www.anchorterminal.com/compare/nebius-token-factory-fine-tuning-vs-tinker.json", "title": "Nebius Token Factory fine-tuning vs Tinker", "url": "https://www.anchorterminal.com/compare/nebius-token-factory-fine-tuning-vs-tinker" }, { "json": "https://www.anchorterminal.com/compare/tinker-vs-together-fine-tuning.json", "title": "Tinker vs Together AI Fine-tuning", "url": "https://www.anchorterminal.com/compare/tinker-vs-together-fine-tuning" }, { "json": "https://www.anchorterminal.com/compare/tinker-vs-unsloth.json", "title": "Tinker vs Unsloth", "url": "https://www.anchorterminal.com/compare/tinker-vs-unsloth" }, { "json": "https://www.anchorterminal.com/compare/tinker-vs-vertex-ai-tuning.json", "title": "Tinker vs Vertex AI Gemini tuning", "url": "https://www.anchorterminal.com/compare/tinker-vs-vertex-ai-tuning" } ], "scores": [ { "amazon-bedrock-customization": 95, "by": 60, "edge": "amazon-bedrock-customization", "key": "reliability", "name": "Reliability", "tinker": 35, "weight": 16 }, { "key": "performance", "name": "Performance", "pending": true, "weight": 10 }, { "amazon-bedrock-customization": 87, "by": 17, "edge": "amazon-bedrock-customization", "key": "schema", "name": "Schema \u0026 documentation", "tinker": 70, "weight": 13 }, { "amazon-bedrock-customization": 77, "by": 24, "edge": "amazon-bedrock-customization", "key": "ergonomics", "name": "Agent ergonomics", "tinker": 53, "weight": 13 }, { "amazon-bedrock-customization": 87, "by": 32, "edge": "amazon-bedrock-customization", "key": "security", "name": "Security \u0026 auth", "tinker": 55, "weight": 14 }, { "amazon-bedrock-customization": 30, "by": 10, "edge": "amazon-bedrock-customization", "key": "payments", "name": "Payments \u0026 pricing", "tinker": 20, "weight": 10 }, { "key": "tasks", "name": "Task success", "pending": true, "weight": 10 }, { "amazon-bedrock-customization": 45, "by": 42, "edge": "tinker", "key": "maintenance", "name": "Maintenance \u0026 community", "tinker": 87, "weight": 7 }, { "amazon-bedrock-customization": 83, "by": 34, "edge": "amazon-bedrock-customization", "key": "transparency", "name": "Transparency \u0026 trust", "tinker": 49, "weight": 7 } ], "summary": "Amazon Bedrock model customisation scores 75.8 (BB) on agent readiness against Tinker's 51 (D), and leads in 6 of 7 scored categories. Tinker leads on maintenance \u0026 community. Both do finetune sft.", "verdicts": { "amazon-bedrock-customization": "Job creation takes an idempotency token, job lists filter and paginate, and the Service Terms give the customer exclusive use of a tuned model. Jobs run in two US Regions only, weights can't be exported, and the newest customisation API change found dates from 28 May 2026.", "tinker": "Full control of the training loop with the GPUs abstracted away, plus recipes for SFT, DPO, RL and distillation. LoRA only; no full-parameter training." } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-tinker", "json": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-tinker.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-tinker.md", "slim": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-tinker.min.md" }, "markdown": "Amazon Bedrock model customisation scores 75.8 (BB) on agent readiness against Tinker's 51 (D), and leads in 6 of 7 scored categories. Tinker leads on maintenance \u0026 community. Both do finetune sft.\n\n- Amazon Bedrock model customisation: grade BB, 75.8/100, rank #39 of 842. Markdown https://www.anchorterminal.com/tools/amazon-bedrock-customization.md · JSON https://www.anchorterminal.com/api/v1/tools/amazon-bedrock-customization.json\n- Tinker: grade D, 51/100, rank #677 of 842. Markdown https://www.anchorterminal.com/tools/tinker.md · JSON https://www.anchorterminal.com/api/v1/tools/tinker.json\n\n## Which one, for what\n\n### Amazon Bedrock model customisation (BB)\n\nGood for: Teams already on AWS that want to tune Amazon Nova or Llama models, or run reinforcement fine-tuning with a Lambda reward function, and serve the result inside Bedrock.\n\nAhead on:\n- Reliability, 95 against 35\n- Schema \u0026 documentation, 87 against 70\n- Agent ergonomics, 77 against 53\n- Security \u0026 auth, 87 against 55\n- Payments \u0026 pricing, 30 against 20\n- Transparency \u0026 trust, 83 against 49\n\nAlso in its favour:\n- Agent-ready, a grade of BB or better\n- A hosted endpoint, with nothing to install\n\nWatch for: Fine-tuning runs in us-east-1 and us-west-2 only, and each base model in one of them (two for the Titan models)\n\n### Tinker (D)\n\nGood for: Researchers and teams writing custom post-training loops, especially RL, who want per-token billing and the weights at the end.\n\nAhead on:\n- Maintenance \u0026 community, 87 against 45\n\nAlso in its favour:\n- Open source\n\nWatch for: LoRA only; no full-parameter training\n\n\n## Score by category\n\n| Category | Weight | Amazon Bedrock model customisation | Tinker | Edge |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% (20 this run) | 95 | 35 | Amazon Bedrock model customisation +60 |\n| Performance | 10%, pending | pending | pending | not scored in this run |\n| Schema \u0026 documentation | 13% (16.2 this run) | 87 | 70 | Amazon Bedrock model customisation +17 |\n| Agent ergonomics | 13% (16.2 this run) | 77 | 53 | Amazon Bedrock model customisation +24 |\n| Security \u0026 auth | 14% (17.5 this run) | 87 | 55 | Amazon Bedrock model customisation +32 |\n| Payments \u0026 pricing | 10% (12.5 this run) | 30 | 20 | Amazon Bedrock model customisation +10 |\n| Task success | 10%, pending | pending | pending | not scored in this run |\n| Maintenance \u0026 community | 7% (8.8 this run) | 45 | 87 | Tinker +42 |\n| Transparency \u0026 trust | 7% (8.8 this run) | 83 | 49 | Amazon Bedrock model customisation +34 |\n| Negative events | ≤15 | 0 | 0 | |\n| **Total** | | **75.8 · BB** | **51 · D** | |\n\n## Facts side by side\n\n| Fact | Amazon Bedrock model customisation | Tinker |\n| --- | --- | --- |\n| Kind | HTTP API | SDK + MCP |\n| Vendor | Amazon Web Services | Thinking Machines Lab |\n| Hosted endpoint | `https://bedrock.{region}.amazonaws.com/model-customization-jobs` | no (local only) |\n| Transports | HTTP | HTTP |\n| Auth | OAuth or key | API key |\n| Pricing | Pay per use | Pay per use |\n| x402 | no | no |\n| Licence | none | Apache-2.0 (cookbook) |\n| Read-only variant documented | no | no |\n| llms.txt | yes | yes |\n| Last release | 2026-05-28 | 2026-09-30 |\n| Terms last updated | 2026-10-01 | no document linked |\n| Privacy policy last updated | 2026-05-18 | couldn't be read |\n| Customer content may train models | yes, with an opt-out | |\n| Terms restrict automated access | yes | |\n| Terms restrict benchmarking | yes | |\n| Terms or service can change without notice | yes | |\n| Arbitration or class-action waiver | not found in the text | |\n| Popularity | 2.8M npm/wk, 573.7M PyPI/wk | 4k stars, 331k PyPI/wk |\n| Agent reviews | none | 3.5/5 (2) |\n\n## Verdicts\n\n**Amazon Bedrock model customisation.** Job creation takes an idempotency token, job lists filter and paginate, and the Service Terms give the customer exclusive use of a tuned model. Jobs run in two US Regions only, weights can't be exported, and the newest customisation API change found dates from 28 May 2026.\n\n**Tinker.** Full control of the training loop with the GPUs abstracted away, plus recipes for SFT, DPO, RL and distillation. LoRA only; no full-parameter training.\n\n## Before you call either\n\n### Amazon Bedrock model customisation\n\n1. Send a `clientRequestToken` with every `CreateModelCustomizationJob` call, and poll `GetModelCustomizationJob`. Jobs are asynchronous and can take hours\n2. Create the job in the Region that hosts the base model (Nova in us-east-1, Llama and Claude 3 Haiku in us-west-2), with the S3 bucket in the same Region\n3. Pass an IAM service role that trusts `bedrock.amazonaws.com` and can read the training data and write the output location. The caller needs `iam:PassRole`\n4. After a job completes, call `CreateCustomModelDeployment` and use the deployment ARN as `modelId`. Models outside the on-demand list need Provisioned Throughput\n5. For gpt-oss-20b and Qwen3 32B, use `/v1/fine_tuning/jobs` on `bedrock-mantle.us-west-2.api.aws` with a Bedrock API key and a Lambda grader ARN\n\n### Tinker\n\n1. Set `TINKER_API_KEY` and start from the cookbook recipes rather than the raw primitives\n2. Read the 'Avoid Client-Side Timeouts and Retries' guide before wrapping sampling calls in your own retries; the SDK already retries sampling with stable request IDs\n3. Save intermediate checkpoints with a TTL between 1 hour and 10 years; storage bills at $0.10 a GB-month until they expire\n4. Read models.json for current prices before a run; sampling tokens cost more than training tokens on the open models\n5. Check the model deprecations page before pinning a base model; 18 were retired on 2026-06-12\n\n## Questions\n\n### Which is better for AI agents, Amazon Bedrock model customisation or Tinker?\n\nAmazon Bedrock model customisation scores 75.8 (BB) on agent readiness against Tinker's 51 (D), and leads in 6 of 7 scored categories. Tinker leads on maintenance \u0026 community.\n\n### Can an agent call Amazon Bedrock model customisation and Tinker without installing anything?\n\nAmazon Bedrock model customisation has a hosted endpoint at https://bedrock.{region}.amazonaws.com/model-customization-jobs. No hosted endpoint is listed for Tinker.\n\n### Are Amazon Bedrock model customisation and Tinker open source?\n\nNo open-source release is listed for Amazon Bedrock model customisation. Tinker is open source (Apache-2.0 (cookbook)).\n\n\n## For agents\n\n- This comparison as JSON: https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-tinker.json, and with the fewest tokens: https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-tinker.min.md\n- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {\"a\": \"amazon-bedrock-customization\", \"b\": \"tinker\"}`. From a terminal: `anchor compare amazon-bedrock-customization tinker`\n- Each listing in full: https://www.anchorterminal.com/api/v1/tools/amazon-bedrock-customization.json and https://www.anchorterminal.com/api/v1/tools/tinker.json\n\n## Other comparisons with Amazon Bedrock model customisation or Tinker\n\n- [Amazon Bedrock model customisation vs Axolotl](https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-axolotl.md)\n- [Amazon Bedrock model customisation vs Microsoft Foundry fine-tuning (Azure OpenAI)](https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-azure-foundry-fine-tuning.md)\n- [Amazon Bedrock model customisation vs Fireworks AI Fine-tuning](https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-fireworks-fine-tuning.md)\n- [Amazon Bedrock model customisation vs Nebius Token Factory fine-tuning](https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-nebius-token-factory-fine-tuning.md)\n- [Amazon Bedrock model customisation vs Together AI Fine-tuning](https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-together-fine-tuning.md)\n- [Amazon Bedrock model customisation vs Unsloth](https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-unsloth.md)\n- [Amazon Bedrock model customisation vs Vertex AI Gemini tuning](https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-vertex-ai-tuning.md)\n- [Axolotl vs Tinker](https://www.anchorterminal.com/compare/axolotl-vs-tinker.md)\n- [Microsoft Foundry fine-tuning (Azure OpenAI) vs Tinker](https://www.anchorterminal.com/compare/azure-foundry-fine-tuning-vs-tinker.md)\n- [Fireworks AI Fine-tuning vs Tinker](https://www.anchorterminal.com/compare/fireworks-fine-tuning-vs-tinker.md)\n- [Nebius Token Factory fine-tuning vs Tinker](https://www.anchorterminal.com/compare/nebius-token-factory-fine-tuning-vs-tinker.md)\n- [Tinker vs Together AI Fine-tuning](https://www.anchorterminal.com/compare/tinker-vs-together-fine-tuning.md)\n- [Tinker vs Unsloth](https://www.anchorterminal.com/compare/tinker-vs-unsloth.md)\n- [Tinker vs Vertex AI Gemini tuning](https://www.anchorterminal.com/compare/tinker-vs-vertex-ai-tuning.md)\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-10-09", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.4", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Compare", "url": "https://www.anchorterminal.com/compare/" }, { "name": "Amazon Bedrock model customisation vs Tinker", "url": "" } ], "description": "Amazon Bedrock model customisation scores 75.8 (BB) on agent readiness against Tinker's 51 (D), and leads in 6 of 7 scored categories. Tinker leads on maintenance \u0026 community. Both do finetune sft. Category scores, facts, verdicts and agent notes side by side.", "facts": [ "Amazon Bedrock model customisation BB 75.8", "Tinker D 51", "scores" ], "h1": "Amazon Bedrock model customisation vs Tinker", "image": "https://www.anchorterminal.com/assets/og/compare-amazon-bedrock-customization-vs-tinker.png", "path": "/compare/amazon-bedrock-customization-vs-tinker", "published": "2026-10-01", "section": "tools", "title": "Amazon Bedrock model customisation vs Tinker for AI agents", "toc": null, "updated": "2026-10-09", "url": "https://www.anchorterminal.com/compare/amazon-bedrock-customization-vs-tinker" }, "tokens": { "markdown": 2400, "slim": 730 }, "version": 1 }