{
  "listings": [
    {
      "slug": "amazon-bedrock-customization",
      "name": "Amazon Bedrock model customisation",
      "company": "Amazon Web Services, Inc.",
      "grade": "A",
      "fit": "caveat",
      "why": "Distillation generates teacher responses from your prompts and tunes the student, and on-demand serving bills per use, but only Nova models and Llama 3.3 70B deploy on demand, each in one Region; others need Provisioned Throughput.",
      "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-distillation.html",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/amazon-bedrock-customization"
    },
    {
      "slug": "azure-foundry-fine-tuning",
      "name": "Microsoft Foundry fine-tuning (Azure OpenAI)",
      "company": "Microsoft Corporation",
      "grade": "C",
      "fit": "caveat",
      "why": "Distillation is documented as SFT on teacher-generated data, but Standard and Global Standard serving add an hourly hosting fee to base-model token rates, and fee-free Developer deployments are removed after 24 hours.",
      "source": "https://learn.microsoft.com/en-us/azure/foundry/fine-tuning/overview",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/azure-foundry-fine-tuning"
    },
    {
      "slug": "fireworks-fine-tuning",
      "name": "Fireworks AI Fine-tuning",
      "company": "Fireworks.ai, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "A distillation cookbook exists and serverless LoRA serves adapters per token at the base price, but only on GLM 5.3 Flash with three adapters per account; adapters on other bases need an on-demand deployment.",
      "source": "https://docs.fireworks.ai/serverless/specialized-model-inference/overview.md",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/fireworks-fine-tuning"
    },
    {
      "slug": "nebius-token-factory-fine-tuning",
      "name": "Nebius Token Factory fine-tuning",
      "company": "Nebius B.V.",
      "grade": "D",
      "fit": "caveat",
      "why": "SFT on chat JSONL suits teacher traces and acc-02 plans precisely this, but tuned models deploy only to dedicated endpoints (custom weights in beta on request) and no per-token price for a tuned model is published.",
      "source": "https://docs.tokenfactory.nebius.com/ai-models-inference/dedicated-endpoints/custom-weights.md",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/nebius-token-factory-fine-tuning"
    },
    {
      "slug": "tinker",
      "name": "Tinker",
      "company": "Thinking Machines Labs, Inc.",
      "grade": "D",
      "fit": "caveat",
      "why": "Cookbook recipes cover on- and off-policy distillation, but per-token serving prices and an evaluation-against-teacher guide sit in the docs, which were not read this time.",
      "source": "https://github.com/thinking-machines-lab/tinker-cookbook",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/tinker"
    },
    {
      "slug": "together-fine-tuning",
      "name": "Together AI Fine-tuning",
      "company": "Together Computer, Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "Conversational JSONL with loss on assistant turns fits teacher traces, but tuned models are served on dedicated endpoints billed per minute even when idle; acc-19 and acc-45 describe serverless LoRA the current docs no longer offer.",
      "source": "https://docs.together.ai/docs/fine-tuning/data-preparation.md",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/together-fine-tuning"
    },
    {
      "slug": "vertex-ai-tuning",
      "name": "Vertex AI Gemini tuning",
      "company": "Google LLC",
      "grade": "B",
      "fit": "caveat",
      "why": "SFT on Flash-Lite from teacher outputs is served per token at the base price in the global region (1.5 times elsewhere for Gemini 3), below a Pro teacher, but no guide to evaluating the student against the teacher was found.",
      "source": "https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/tuning.md.txt",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/vertex-ai-tuning"
    },
    {
      "slug": "axolotl",
      "name": "Axolotl",
      "company": "Axolotl AI",
      "grade": "B",
      "fit": "no",
      "why": "Trains a student from chat datasets, but serves nothing, so the per-token price and side-by-side switch the success needs lie outside the tool.",
      "source": "https://docs.axolotl.ai/docs/dataset-formats/conversation.html",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/axolotl"
    },
    {
      "slug": "unsloth",
      "name": "Unsloth",
      "company": "Unsloth",
      "grade": "C",
      "fit": "no",
      "why": "It can train a student on teacher traces, but it serves nothing, so the per-token price and swap-over the success needs lie outside the tool.",
      "source": "https://unsloth.ai/docs/basics/inference-and-deployment/vllm-guide.md",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/unsloth"
    }
  ],
  "marks": "yes: the docs show it does the job as the success test says. caveat: it does the job with a limit an agent should know. no: it doesn't do this job. unchecked: we couldn't read it.",
  "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": "2026-11-16",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.8",
    "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"
  },
  "useCase": {
    "accounts": 6,
    "after": [
      "inference.fast-small-calls",
      "open-weights.sub-task-cousin"
    ],
    "area": "models",
    "asks": [
      "make a cheaper model that does this step as well",
      "distil our judge into a small model",
      "train a small model on GPT outputs for this task"
    ],
    "before": [
      "agent-observability.coding-agent-sessions",
      "agent-observability.offline-eval-regression"
    ],
    "blockers": [
      "tuned model served only on a dedicated endpoint, so it costs more than the call it replaces",
      "traces need cleaning the service does not do"
    ],
    "capabilities": [
      "finetune.sft",
      "finetune.distill",
      "finetune.lora"
    ],
    "category": "fine-tuning",
    "categoryName": "Fine-tuning",
    "checked": "2026-11-15",
    "companies": {
      "fit": 0,
      "fitWithLimit": 7,
      "tagged": 13
    },
    "competition": "none",
    "fits": null,
    "fitsWithLimit": [
      "amazon-bedrock-customization",
      "azure-foundry-fine-tuning",
      "fireworks-fine-tuning",
      "nebius-token-factory-fine-tuning",
      "tinker",
      "together-fine-tuning",
      "vertex-ai-tuning"
    ],
    "id": "fine-tuning.distil-to-small",
    "listings": {
      "caveat": 7,
      "no": 2,
      "notRead": 0,
      "tagged": 13,
      "unchecked": 0,
      "yes": 0
    },
    "name": "Replace an expensive model call with a small tuned model",
    "status": "draft",
    "success": "Teacher outputs or traces can be turned into a training set in the documented format, the tuned small model is served at a published per-token price below the model it replaces, and the docs show how to evaluate it against the teacher on held-out cases before switching.",
    "url": "https://www.anchorterminal.com/api/v1/use-cases/fine-tuning.distil-to-small.json",
    "weight": 12,
    "weightBasis": "6 of 47 accounts: an agent picking and tuning a replacement model, a judge distilled on Token Factory, a Fireworks tune for consistent output, an email sorter, a forum design for a trace-trained coding adapter and one vendor benchmark"
  }
}
