Kimi K2.7 Code

Released2026-06-12
Price per 1M$0.95 in · $4.00 out
Overall score55.4 · rank 72
Context window262,144 tokens
ProviderMoonshot AI
WeightsOpen weights

Scores

DimensionScoreRankBenchmarks
Overall: Balanced 55.4 72 20
Overall: Coding 55.2 68 16
Overall: Reasoning 55.4 71 8
Overall: Agentic work 56.6 64 14
Overall: Frontend 52.7 73 8
Coding 55.3 69 6
Agents 57.2 59 4
Reasoning 56.1 76 4
Math 56.5 59 3
Knowledge 52.0 81 1
Frontend 51.5 71 2
Value vs price+6.163

Configurations

ConfigurationBalancedBenchmarks
Unspecified setting 55.2 ±1.4 19
Non-reasoning 56.2 ±5.1 1

Evidence

Coding 55.3 ±1.4

BenchmarkResultSourceDate
FrontierCode 30.1% Epoch AI 2026-11-11
DeepSWE 30.5% Epoch AI 2026-11-11
SWE-bench Verified (Vals run) 78.2% Vals AI 2026-09-01
SciCode 47.8% Epoch AI 2026-11-11
LiveCodeBench (Vals run) 82.1% Vals AI 2026-09-01
ProgramBench 0.0% Vals AI 2026-11-09

Agents 57.2 ±2.5

BenchmarkResultSourceDate
APEX-Agents 37.6% Epoch AI 2026-11-11
Terminal-Bench 2.1 (Vals run) 67.0% Vals AI 2026-10-01
SkillsBench 50.0% Vals AI 2026-10-29
Vending-Bench 2 $5,083 Epoch AI 2026-11-11

Reasoning 56.1 ±3.0

BenchmarkResultSourceDate
GPQA Diamond 87.9% Epoch AI 2026-08-07
CritPt 10.0% Epoch AI 2026-11-11
Chess Puzzles 14.0% Epoch AI 2026-10-13
SimpleBench 57.9% SimpleBench 2026-07-09
SimpleBench 57.9% Epoch AI 2026-11-11

Math 56.5 ±2.0

BenchmarkResultSourceDate
FrontierMath Tiers 1-3 54.0% Epoch AI 2026-06-13
FrontierMath Tier 4 12.2% Epoch AI 2026-06-13
OTIS Mock AIME 2024-2025 95.6% Epoch AI 2026-08-07

Knowledge 52.0 ±3.0

BenchmarkResultSourceDate
SimpleQA Verified 36.5% Epoch AI 2026-08-27

Frontend 51.5 ±0.7

BenchmarkResultSourceDate
Vibe Code Bench v1.1 47.2% Vals AI 2026-11-09
LMArena WebDev 1,474 LMArena 2026-11-09

Names matched to this model

kimi-k2-7-code

Compare Kimi K2.7 Code with other models