Claude Fable 5

What is Kimi K3?#

Kimi K3 is Moonshot AI's flagship language model, released on July 16, 2026, built for long coding runs, agentic web research, and document work. It is a from-scratch architecture rather than a K2 revision: 2.8 trillion parameters with 16 of 896 experts active per pass, and a hybrid linear attention design called Kimi Delta Attention paired with Attention Residuals. That attention work is the reason the 1M-token window is usable rather than decorative. On Unifically it runs as moonshotai/kimi-k3 with a 1,048,576 token context, 131,072 output tokens by default, and thinking on every request.

Key features of Kimi K3#

Top-10 on Arena Text

K3 ranks #10 on Arena's Text board at 1486±10, between Gemini 3 Pro at 1486 and GPT 5.6 Sol (xhigh) at 1485. It is the highest-ranked model on that board from outside the big closed labs, out of 378 ranked models. The rating is still marked preliminary at 3,619 votes and has already moved several places since launch week, so expect it to keep moving.

A 29-place jump in one generation

Kimi K2.6 sits at #39 with 1461 Elo. K3 lands at #10 with 1486, a 25-point gain and 29 places in a single release, while the context window goes from 262k to 1M tokens.

1M token context, 131k default output

The window holds a large repo, a long filing pack, or a full agent transcript at once. A single response defaults to 131,072 tokens and can be raised to the full 1M when you need one long artifact instead of many calls.

Full agent and tool surface

K3 exposes the whole tool-calling surface through the same endpoint: tool definitions, `tool_choice` including `required`, dynamic tool loading for agents carrying a large tool set, strict `json_schema` structured output, and partial mode. Streaming returns reasoning and answer text as separate deltas, so you can show thinking without parsing it out of the response.

Fourth on the Artificial Analysis index

Independent testing puts K3 at #4 of 187 models on the Intelligence Index at a score of 57. That is the strongest outside confirmation of the launch claims, since most of the benchmark table Moonshot published has not been replicated yet.

Weights promised by July 27, 2026

Moonshot committed to a public weights release and a technical report by July 27. Nothing has been published yet. If it lands, K3 becomes the largest open model by a wide margin, though 2.8T parameters means a multi-node deployment, not a workstation.

Best for#

Use cases#

Build a coding agent that plans a migration, edits files across a repo, runs tests, and keeps going across a long session without restarting. Use it as the research layer in a product that has to plan queries, read what comes back, and synthesise an answer with sources. It also fits document intelligence work where long contracts, filings, and spreadsheets are read together and returned as structured output. For teams already running an OpenAI-compatible agent loop, K3 slots in as a model id change.

Limitations#

Thinking is always on and reasoning_effort accepts only max. Every request pays full reasoning cost, so simple chat and extraction carry latency and token spend they do not need. Lower effort modes are promised but not shipped.

Speed is a real constraint. Artificial Analysis measures 62 output tokens per second, below the median for models in its class.

The model is sensitive to thinking history. Moonshot trained it in a preserved-history mode, and a harness that fails to pass earlier reasoning content back, or a session switched to K3 from another model mid-run, can produce unstable output.

It also acts rather than asks. On ambiguous instructions K3 makes unrequested decisions, so constraints belong in the system prompt rather than in follow-up corrections.

Moonshot says plainly that overall performance still trails Claude Fable 5 and GPT 5.6 Sol, and most of the benchmark table it published uses its own in-house harness while competitor rows use other vendors' tooling. Treat those self-reported numbers as unconfirmed.

Kimi K3 vs Claude Fable 5#

On Arena's Text board the two are closer than anywhere else: K3 sits at #10 with 1486±10 against Fable 5 at #1 with 1507, a 21-point gap, and K3's rating is still preliminary at 3,619 votes. The separation shows up elsewhere. Fable 5 runs adaptive thinking, so it spends reasoning only where a task needs it, while K3 runs max effort on every call. Fable 5 also takes image input alongside text through the same endpoint. Moonshot itself puts Fable 5 ahead on overall user experience. Pick K3 for long autonomous coding and web research runs where token cost per unit of work matters; pick Fable 5 when you want the stronger all-round model and control over how much it thinks.

When to use Kimi K3#

Use Kimi K3 for work that runs long and mostly unattended: repo-scale coding, multi-step research, and document passes that would otherwise be split across many prompts. It is a poor fit for chat, short extraction, and anything latency-sensitive, because it thinks at full effort every time and generates slower than most models in its class.

API examples#

Call Kimi K3 by POSTing to /v1/chat/completions, the OpenAI-compatible endpoint shared by every language model on the platform. Parameter docs live at docs.unifically.com/models/llm/overview.

curl -X POST https://api.unifically.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "moonshotai/kimi-k3",
    "messages": [
      { "role": "user", "content": "Plan a migration from Express to Fastify for this service and list the risks in order." }
    ]
  }'

The response comes back synchronously with the completion. Set "stream": true to receive tokens as they generate; thinking and answer text arrive as separate deltas.