Model

Kimi K2 Thinking

OPEN

Moonshot AI · Kimi K2 family · released Nov 6, 2025

Open-weight 1T MoE reasoning agent with native INT4 weights and 256K context, chaining 200-300 sequential tool calls.

ReasoningCodingVisionFunction callingTool useAgentic

2161.6

Elo · rank #64

Parameters

1000B

Active params

32B (MoE)

Context

256K tokens

Architecture

Mixture-of-Experts, 384 experts (8 active + 1 shared/token), native INT4 quantization-aware training

License

Modified MIT License

Languages

—

API price (in/out)

$0.6 / $2.5

Modalities

text

Benchmark results

Bar shows position within the tracked field; marker = field best

AIMEMath99.1%#6

best: GPT-5.2 · 100.0%

BrowseCompAgents60.2%#30

best: Kimi K3 · 91.2%

GPQA DiamondReasoning84.5%#64

best: GPT-6 Astra · 96.0%

Humanity's Last ExamReasoning44.9%#15

best: Claude Opus 5 · 64.7%

IFBenchReasoning68.0%#32

best: MiniMax M3 · 83.0%

LiveCodeBenchCoding83.1%#36

best: DeepSeek-V4-Pro (Think Max) · 93.5%

MMLU-ProKnowledge84.6%#30

best: Claude Fable 5 · 91.5%

MMLU-ReduxKnowledge94.4%#5

best: Qwen3.7-Max · 95.0%

SWE-bench VerifiedCoding71.3%#58

best: Claude Opus 5 · 96.0%

τ²-Bench TelecomAgents93.0%#17

best: Claude Opus 4.6 · 99.3%

Terminal-BenchCoding47.1%#2

best: Claude Opus 4.5 (High) · 59.3%

Run it locally

VRAM @ Q4

600 GB

VRAM @ FP16

2000 GB

Fits on (Q4)

Multi-node cluster required

Does not fit on a single RTX 4090 (24GB); native INT4 packs ~500-600GB combined RAM/VRAM, community reports only ~5-10 tok/s with heavy CPU/SSD offload

Quantizations

GGUF · INT4 (native QAT)

Fine-tune it

Permissive

QLoRA680.0 GB4× B200 192GB

LoRA2130.0 GBbeyond 8× B200

Full fine-tune16050.0 GBbeyond 8× B200

QLoRA SFT on ~10k samples ≈ $11.44 (4× B200 192GB)

Plan a fine-tune →

Kimi K2 family

Elo progression across releases

Kimi K2Jul 20251491.9 Kimi K2 ThinkingNov 20252161.6 Kimi K2.5Jan 20262233.3 Kimi K2.6Apr 20262471.3 Kimi K2.7 CodeJun 20262223.5

Resources

HFmoonshotai/Kimi-K2-Thinking GHmoonshotai/Kimi-K2 DOCSmoonshotai.github.io/Kimi-K2/thinking.html APIAvailable via Moonshot AI API

API price $0.6/$2.5 · each benchmark row carries its own source badge (see methodology)