#02Moonshot· released 2026-07-16

MoonshotAI: Kimi K3.

Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...

View on OpenRouter ↗
Context
1.0M
Input $/1M
$3.00
Output $/1M
$15.00
Modality
text · image

Benchmarks

How these are aggregated →
Arena Elo
1558
LMArena — human blind votes
AA Index
71.0
Artificial Analysis composite
LiveBench
83.8
Contamination-resistant
GPQA
90.4%
Grad-level science reasoning
MMLU-Pro
90.9%
Broad knowledge
SWE-bench
80.3%
Real GitHub bugs (Verified)
Terminal-Bench
57.2%
Agentic shell tasks
ARC-AGI
22.7%
Abstract reasoning
Aider Polyglot
86.9%
Multi-language code editing
Speed
55 tok/s
Median output speed

Nearby on the leaderboard