Back to Models
moonshotai/kimi-k2
Not Available
Kimi K2 0711
Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. Kimi K2 excels across a broad range of benchmarks, particularly in coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) tasks. It supports long-context inference up to 128K tokens and is designed with a novel training stack that includes the MuonClip optimizer for stable large-scale MoE training.
7/11/2025
131,000 tokens
#52 Vision (Overall)
Specifications
Modalities
Input
text
Output
text
Supported Parameters
frequency_penalty
logprobs
max_tokens
min_p
presence_penalty
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_logprobs
top_p
Frequently asked questions
- Is Kimi K2 0711 available on FastMetal?
- Not at the moment. Kimi K3, from the same lab, is available on the FastMetal API today.
- What is the context window of Kimi K2 0711?
- 131,000 tokens, shared between the prompt and the response.
- How does Kimi K2 0711 rank?
- #105 on the public arena's Japanese board (ELO 1,365). Ranks move as the leaderboard is updated.
Leaderboard
Text
OverallELO: 1,418
#119JapaneseELO: 1,365
#105ChineseELO: 1,454
#117KoreanELO: 1,341
#133EnglishELO: 1,420
#137frenchELO: 1,419
#131germanELO: 1,413
#103spanishELO: 1,383
#147russianELO: 1,410
#113CodingELO: 1,460
#126MathELO: 1,388
#163Creative WritingELO: 1,370
#142Instruction FollowingELO: 1,380
#162Hard PromptsELO: 1,432
#128Multi-TurnELO: 1,404
#142Vision
OverallELO: 1,237
#52Code
OverallELO: 1,323
#105