Kimi K2.6
moonshot/kimi-k2.6- Moonshot AI
- Family · Kimi K2
- Released 20 Apr 2026
- Open weights
- Text · Image → Text
- Tools · JSON · Vision
Spec & list price
Context256K
Max output236K
Input / 1M$0.65
Output / 1M$3.415.2× input
Scores · standing
Intelligence27.0#78/489
Coding61.8#38/196
Usage
Tokens / day19.4B▲ 1.7% 7d
Usage rank#74
Requests / day1.1M
Tool calls / req0.29
Per request · 30 days
Avg input16.4K1.1× median
Avg output9391.4× median
Cache hit68%about median
Reasoning share66%1.2× median
Tool-call failures1.1%0.6× median
Consumption
Peak 32.1B 28 SepLow 11.5B 4 OctRank today #74
News & trends
2 mentions this week ▼ 82%this model · videos, posts, articles · 30 days
Trending · 14 days
- Anthropic Haiku 5.5 vs Moonshot Kimi K2.6 draw “a beach scene” #shorts199 views
- Google Gemini 3.5 Flash-Lite vs Moonshot Kimi K2.6 draw “werewolf at pumpkin party” #shorts85 views
- Which Kimi K2.6 API provider actually fits your workload?2 upvotes
News · latest from Moonshot AI
- ReplaySSM: Rethinking State Storage for Faster Hybrid-Model DecodingNVIDIA · 12 Oct
- Together Link: open models in the harness you already use. Start with one command today.Together AI · 5 Oct
- Kimi K3: the complete developer guideTogether AI · 1 Aug
- Together AI announces strategic partnership with Moonshot AI to natively serve Kimi modelsTogether AI · 29 Jul
Benchmarks
bar = share of the top score · Benchmark data: Artificial AnalysisSpeed · median across 16 providers, latest day
Output speed#286/33435 tok/s1.9× slower than median
Fastest provider73 tok/sbest of 16 providers
First token#1861.2 s1.2× longer than median
#13 cheapest of 24 with a similar score
Value
Intelligence Index
Blended price per 1M tokens (log)
Family progress
Intelligence Index
- Family releases27.0
Technical details
- Tokenizer
- Other
- License
- other
- Parameters
- 1T
- Moderation
- none
- Downloads · likes
- 249.8K · 1.6K
Request parameters 20
- tools
- tool_choice
- parallel_tool_calls
- response_format
- structured_outputs
- reasoning
- include_reasoning
- temperature
- top_p
- top_k
+10 more
- min_p
- max_tokens
- stop
- seed
- frequency_penalty
- logit_bias
- logprobs
- presence_penalty
- repetition_penalty
- top_logprobs
Interesting facts
- The most used of 5 Kimi K2 models by tokens last month.
- Released 196 days ago, the newest of 5 Kimi K2 models.
When to use
Best for
- Agentic tool-calling workflows
- Instruction-following assistants
- Science question answering
- Coding help in image-aware tools
- Self-hosted deployments on open weights
Think twice for
- Video input
- Image generation
- New projects (usage is declining)
Research papers
- The LLENS Platform for On-Premises Large Language Model Agents in a Japanese University Hospital: Development and Evaluation Study23 Oct
- AI Mindmeld: A Latent Interconnect Protocol21 Oct
- Training Natural Schemers: RL Can Amplify Oversight-Conditional Behavior19 Oct
- Memory Echo in Single-Agent Long-Term Memory: Lifecycle Evaluation and Provenance-Guided Mitigation13 Oct
- Dual- versus Single-Suggestion AI Support for Radiographic Interpretation in Residents: Randomized Multireader Study7 Oct
- Fitting Vision Adapters at Frontier Scales5 Oct
- Bidirectional Preference Synthesis: Learning Prompt-Conditioned Preferences from Boundary Failures3 Oct
- EASE: Behavior-Adaptive Skill Curation for Self-Evolving Agents29 Sep
Same family
Kimi K2- 22.0
- 23.5
- 12.7
- 15.3
Alternatives
other vendors · similar capability- 29.2
- 26.3
- T25.0
- 22.2
Using Kimi K2.6 at volume?
Sign in to ask for a volume discount. You will continue in the OSMEXO dashboard.








