The Open-Source AI Stack
RSS
All models

Models · kimi

Kimi K2 Instruct

Open Moonshot AI · 2025-07-11 · Modified MIT

A trillion-parameter open-weights MoE optimized for agentic tool-use, with strong SWE-Bench results making it a viable open alternative to closed coding agents at release.

Cost

— / Mtok input
— / Mtok output

Moonshot API · as of 2026-05-19

via Artificial Analysis ↗

Speed

— tok/sec output

Moonshot API · as of 2026-05-19

via Artificial Analysis ↗

Architecture

tokens in Embedding vocab not disclosed · kimi tokenizer × N layers Multi-head Latent Attention RoPE context 128,000 tokens MoE Router 384 experts total · 8 active per token shown: 32 of 384 Output projection tokens out
Schema-generated from data/models.yaml. Every label is auditable against the model's sources.

Specs

Architecture
moe
Total params
1T
Active params
32B
Experts
384 total · 8 active
Context window
128K tokens
Attention
mla
Position encoding
rope
Pretraining tokens
15.5T
Post-training
sft, rlhf
OSI-approved
no
Data released
no
Training code
not released

Benchmarks

Each score carries the date it was published; we never infer or interpolate missing scores.

General reasoning

MMLU-Pro 81.1 as of 2025-07-15 source ↗
GPQA-Diamond 75.1 as of 2025-07-15 source ↗

Code

SWE-Bench Verified 65.8 as of 2025-07-15 source ↗

Recommended use cases

  • agentic tool use
  • code agent backend
  • long-context tasks

Available quantizations

GGUF llama.cpp's container; the common local format, k-quants from Q2 to Q8. runs on llama.cpp, Ollama
GPTQ Post-training 4-bit weight quantization for GPU serving. runs on vLLM, SGLang, Transformers
MLX Apple MLX 4/8-bit layout for Apple silicon. runs on Apple MLX
FP8 8-bit float, frequently a native release on Hopper / Blackwell GPUs. runs on vLLM, SGLang, TensorRT-LLM
bitsandbytes On-the-fly NF4 / INT8 weight quantization inside Transformers. runs on Transformers

Verified via the Hugging Face model tree ↗. Community quantizations change over time; the families shown are those with published weights at audit time.

Notable innovations

  • · 1T-param open-weights MoE
  • · Agentic post-training focus

Lineage

Moonshot's first widely-released open-weights MoE.

Sources