MMODELYST
Models/Kimi K2.7 Code
MDL

Kimi K2.7 Code

Kimi

Verified against Artificial Analysis · Jul 13, 2026
Capability
88
percentile index
Price /M
$0.95 / $4
in / out
Context
262.1K
tokens
Avg score
62.0
7 benchmarks
AA Intelligence
41.9
index
Speed
46 tok/s
TTFT 1.2s
Released
Jun 12, 2026
About

Kimi K2.7 Code is a open-weights model in the Kimi family from Kimi. Benchmarked on 7 evals, averaging 62.0.

Benchmark scores · 7
avg 62.0 — higher is better
τ²-bench
90.1
GPQA Diamond
89.6
AA Long-Context Reasoning
66.3
IFBench
63.1
SciCode
47.5
Terminal-Bench Hard
44.7
Humanity's Last Exam
32.8
Details
PublisherKimi
FamilyKimi
Parameters
License
WeightsOpen
ReleasedJun 12, 2026
Performance & price
Output speed46 tok/s
Latency (TTFT)1.2s
First answer50s
Price in / out$0.95 / $4 per 1M
Blended price$1.71 / 1M
Coding index60.8

Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →

Capability over time
AA Long-Context Reasoningfirst seen at 66.3
24d ago
Terminal-Bench Hardfirst seen at 44.7
24d ago
τ²-benchfirst seen at 90.1
24d ago
IFBenchfirst seen at 63.1
24d ago
SciCodefirst seen at 47.5
24d ago
Humanity's Last Examfirst seen at 32.8
24d ago

Append-only ledger — every observed change to this model's numbers.

API providers
Artificial Analysis$4/M