Kimi K2.6
The prior Kimi generation, a capable and more affordable open-weights model for agentic and long-context tasks. A practical choice when K3's full strength isn't required.
At a glance
Released
Jul 14, 2026
Provider
Moonshot AI
Context window
256,000 tokens
Max output
64,000 tokens
Modalities
Text + image → text
Knowledge cutoff
March 2026
API model ID
kimi-k2.6License
Modified MIT
Pricing
Standard
Input$0.95 / 1M
Output$4 / 1M
* Rates are from a third-party aggregator — verify before relying on them. A typical 1,500-token prompt with a 650-token answer costs ≈ $0.004 on the standard tier. Run your own prompt through the calculator →
What's new vs the previous model
Compared withKimi K2 (January 2026)
- Better tool use and reasoning than the original K2
- Shorter context and weaker planning than K3
Best use cases
Cost-aware agents
Runs multi-step tool workflows at a lower price than K3.
Long-context tasks
Handles large documents and codebases with a wide window.
Self-hosting
Deployable privately under its open license.
Strengths & limitations
Strengths
- Open weights
- Good agentic ability
- Lower cost than K3
Limitations
- Less capable than K3
- Shorter context window
How to test it
- No-code: open kimi.com and pick Kimi K2.6 from the model switcher — no API key needed.
- Playground: sign in to the Moonshot AI Platform and run prompts in the workbench, then copy the generated code.
- API: create a key in the Moonshot AI Platform and call the API (example below).
curl https://api.moonshot.ai/v1/chat/completions \
# OpenAI-compatible endpoint
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "kimi-k2.6",
"messages": [{"role": "user", "content": "Explain tokenization in one paragraph."}]
}'