Kimi K2 0905
Kimi K2 0905Rankingin non-thinking mode?
Alternatives?
CheaperGPT OSS 120BhighOpen+63%-93%$0.11Context: 256K → 131KBetterGPT OSS 20BhighOpen+56%-96%$0.011Context: 256K → 131K
OpenDeepSeek V3.2 Specialethinking+69%——Context: 256K → 128K
CheaperNemotron 3 Nano 30B A3BthinkingOpen+59%-87%$0.017Context: 256K → 131K
CheaperQwen3 235B A22B Thinking 2507thinkingOpen+59%-7%$0.08No strict JSONGPT OSS 120BhighOpen+63%-93%$0.11Context: 256K → 131KNemotron 3 Nano 30B A3BthinkingOpen+59%-87%$0.017Context: 256K → 131KQwen3 235B A22B Thinking 2507thinkingOpen+59%-7%$0.08No strict JSONGPT-5 Minihigh+58%-14%$0.053DeepSeek V3.1thinkingOpen+56%-47%$0.25 / $0.95Context: 256K → 128KGPT OSS 20BhighOpen+56%-96%$0.011Context: 256K → 131KNemotron 3 Nano 30B A3BthinkingOpen+59%-87%$0.017Context: 256K → 131KGPT-5 Minihigh+58%-14%$0.053Qwen3 235B A22B Thinking 2507thinkingOpen+59%-7%$0.08No strict JSONGPT OSS 120BhighOpen+63%-93%$0.11Context: 256K → 131KDeepSeek V3.2 SpecialethinkingOpen+69%——Context: 256K → 128K
GLM 4.7thinkingOpen+66%+25%$0.60 / $2.20Context: 256K → 203KKimi K2 ThinkingthinkingOpen+65%+7%$0.47 / $2GPT OSS 120BhighOpen+63%-93%$0.11Context: 256K → 131KNemotron 3 Nano 30B A3BthinkingOpen+59%-87%$0.017Context: 256K → 131KKimi K2 ThinkingthinkingOpen+65%—$0.47 / $2Kimi K2 0711non-thinkingOpensame—$0.50 / $2Context: 256K → 128K+1Kimi K3.1 Prohighn/a—$2 / $10Kimi K3.1 Pro Highspeedn/a—$4 / $20Kimi K3maxOpenn/a—$2.00Memory (4-bit): 616 GB → 1,668 GB+4Kimi K2.7 CodethinkingOpenn/a—$0.54Instant mode is not supported · Self-hosted / vLLM / SGLangKimi K2.6thinkingOpenn/a—$0.71Kimi K2.5thinkingOpenn/a—$0.60 / $3All Moonshot AI models →Run it yourself
estimateHistory
5 versionsBenchmarks
all categoriesFAQ
How much does Kimi K2 0905 cost?›
$0.40 per 1M input tokens and $2 per 1M output tokens (cheapest host: DeepInfra).
Is Kimi K2 0905 free to use?›
There is no free API; the cheapest paid price is $0.40 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Kimi K2 0905, and what is it best at?›
#239 of 336 in Chat & Knowledge, #81 of 239 in Coding, #90 of 242 in Agents & Tool Use, #49 of 96 in Writing & Language.
Is Kimi K2 0905 better than Kimi K2 0711?›
On AIME 2025 · AA archived, Kimi K2 0905 scores 57.3 vs 57 — about the same.
How can I use Kimi K2 0905?›
Through the API of DeepInfra, Featherless, Novita.
What is the context window of Kimi K2 0905?›
256K tokens, with up to 98K output tokens.
When was Kimi K2 0905 released? Is there a newer version?›
Released 2025-09-05. It has been superseded by Kimi K2.5.
Is Kimi K2 0905 open source?›
Yes, the weights are open (modified-mit licence).
Can I run Kimi K2 0905 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Kimi K2 0905 read images?›
No, it accepts text only.
What is Kimi K2 0905?›
Kimi K2 0905 is an open-weights language model from Moonshot AI, released Sep 2025, the successor to Kimi K2 0711.
What are the best alternatives to Kimi K2 0905?›
GPT OSS 120B (cheaper), GPT OSS 20B (better), DeepSeek V3.2 Speciale (open weights).