Kimi K2.7 Code
Kimi K2.7 CodeRankingin thinking mode?
Alternatives?
CheaperGPT-6.1 Solxhigh+54%-27%$0.39Remove unsupported parameters+5BetterGPT-6 Lunamedium+9%-97%$0.017Function calling only with none · Chat Completions+3
OpenDeepSeek V4 Pro 0813max+24%+25%$0.67Memory (4-bit): 616 GB → 990 GB+6CheaperGPT-5.6 Solmedium+46%-7%$0.50
CheaperMiniMax M3.1 Flashthinking+42%-59%$0.22Complete model response must be appendedGPT-6.1 Solxhigh+54%-27%$0.39Remove unsupported parameters+5GPT-5.6 Solmedium+46%-7%$0.50MiniMax M3.1 Flashthinking+42%-59%$0.22Complete model response must be appendedGPT-6 Solhigh+36%-31%$0.37
Gemini 3 Flash Previewthinking+35%-34%$0.50 / $3GPT-6 Lunamedium+9%-97%$0.017Function calling only with none · Chat Completions+3GPT-5.6 Lunahigh+6%-92%$0.044GPT-6.1 Sollow+49%-76%$0.13Remove unsupported parameters+5GPT-6 Sollow+30%-75%$0.13GPT-5.6 Terralow+11%-73%$0.14Prompts with >272K input tokensDeepSeek V4 Pro 0813maxOpen+24%+25%$0.67Memory (4-bit): 616 GB → 990 GB+6DeepSeek V4.1 Flash 1017maxOpen+8%-52%$0.26Pass back reasoning_content · Chat Completions+4TInklingxhighOpen+5%same$0.95 / $4.05No strict JSON
Step 5 PreviewthinkingOpen+5%+90%$1.03web_search not available · Chat Completions+2
Muse Spark OpenxhighOpensame+56%$0.84Reasoning off leaves output unchanged · Self-hosted / llama.cpp+1Kimi K3maxOpen+20%+270%$2.00Memory (4-bit): 616 GB → 1,668 GB+4Kimi K2.5thinkingOpen-11%—$0.60 / $3Kimi K2.6thinkingOpen-18%+48%$0.80Kimi K2 ThinkingthinkingOpen-22%—$0.47 / $2No image inputKimi K2 0711non-thinkingOpen-31%—$0.50 / $2Context: 256K → 128K+2Kimi K2 0905non-thinkingOpen-36%—$0.40 / $2No image inputKimi K3.1 Pron/a—$2 / $10Kimi K3.1 Pro Highspeedn/a—$4 / $20All Moonshot AI models →Run it yourself
estimateBenchmarks
all categoriesWatch out 1
Instant mode is not supported · Self-hosted / vLLM / SGLang›
The documented third-party vLLM and SGLang deployments do not support Instant mode. Source ↗
FAQ
How much does Kimi K2.7 Code cost?›
$0.95 per 1M input tokens and $4 per 1M output tokens; cached input $0.19 per 1M; cheapest via Inceptron at $0.67 / $3.35. A typical benchmark task costs about $0.54 at default effort.
Is Kimi K2.7 Code free to use?›
There is no free API; the cheapest paid price is $0.67 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Kimi K2.7 Code, and what is it best at?›
#62 of 334 in Chat & Knowledge, #54 of 132 in Coding, #54 of 133 in Agents & Tool Use, #77 of 245 in Writing & Language, #52 of 133 in Vision & Documents. Strongest at: Tool calling, Reasoning, Science, Long context.
Is Kimi K2.7 Code better than Claude Sonnet 5?›
On AA Omniscience · Accuracy, Kimi K2.7 Code scores 39.6 vs 40.1 — about the same. It is 89% cheaper per task.
How can I use Kimi K2.7 Code?›
Through the API of Alibaba, Cloudflare, CoreWeave, DeepInfra, Featherless and 8 more.
What is the context window of Kimi K2.7 Code?›
256K tokens, with up to 236K output tokens.
When was Kimi K2.7 Code released? Is there a newer version?›
Released 2026-06-12. It is the current version in its line.
Is Kimi K2.7 Code open source?›
Yes, the weights are open (modified-mit licence).
Can I run Kimi K2.7 Code locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Kimi K2.7 Code read images?›
It accepts text and images as input.
What is Kimi K2.7 Code?›
Kimi K2.7 Code is an open-weights language model from Moonshot AI, released Jun 2026.
What are the best alternatives to Kimi K2.7 Code?›
GPT-6.1 Sol (cheaper), GPT-6 Luna (better), DeepSeek V4 Pro 0813 (open weights).