Kimi K2 Instruct kimi-k2-instruct

moonshotai · kimi flagship agentic

#27 overall #2 coding #4 agentic

params
1000B (A32B)
arch
moe
context
128k
license
modified-mit open weights
released
Jul 2025
train compute
3.0e+24 FLOP
downloads/30d
336.7k

# why ranked

Overall: #27 — score 28.2 ● medium — one signal missing, weights renormalized

signalweightinput (0–100)
aa_intelligence 0.50 19.4
arena_elo 0.30
bench_composite 0.20 50.2

missing signals are dropped and the remaining weights renormalized — never imputed.

Coding: #2 — score 76.1 ● medium — one signal missing, weights renormalized

signalweightinput (0–100)
aa_coding 0.25
aider_polyglot 0.25 72.7
arena_elo_code 0.15
swe_bench_verified 0.35 78.6

missing signals are dropped and the remaining weights renormalized — never imputed.

Agentic: #4 — score 78.6 ● low — a single signal; treat with caution

signalweightinput (0–100)
swe_bench_verified 1.00 78.6

full methodology

# trend

OGM score · last 2 days
OGM score over 2 days
overall rank (up is better)
Overall rank over 2 days

# benchmarks

benchmarkscoresourcedate
AA Intelligence Index 19.4 Artificial Analysis
AA Math Index 57.0 Artificial Analysis
Aider Polyglot 59.1 Aider Jul 2025
Acebench 0.8 / 1 LLM Stats
Aider-polyglot 0.6 / 1 LLM Stats
AIME 2024 0.7 / 1 LLM Stats
AIME 2025 0.5 / 1 LLM Stats
Autologi 0.9 / 1 LLM Stats
Cnmo-2024 0.7 / 1 LLM Stats
Csimpleqa 0.8 / 1 LLM Stats
Gsm8k 1.0 / 1 LLM Stats
Math-500 1.0 / 1 LLM Stats
SWE-bench Verified 65.4 SWE-bench Jul 2025

# where to run

providerquantctx$/M in$/M out$/M blendtpsuptime
Novitafp8 128k $0.57$2.30$1.00 99.9%

blended = (3·input + output) / 4 per 1M tokens · cheapest first

# variants

Kimi K2 Instruct 0905 Kimi K2 Thinking

source aliases
aa
kimi-k2
aider
Kimi K2
arena
Kimi-K2-Instruct
epoch
Kimi K2
openrouter
moonshotai/kimi-k2
swebench
moonshot/kimi-k2-0711-preview