BenchLeader
KimiOpen weightsAuto-detected

Kimi Linear 48B A3B Instruct

Not yet ranked: too few independent results so far.

Blended price
Output speed
First answer
Context
1M
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Agents & tools43
  2. Instruction following32
  3. Long context39
  4. Composite38

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

BenchmarkdefaultSource
GPQA Diamond (AA)not in index41.2%#448Artificial Analysis
Humanity's Last Exam (AA)not in index2.5%#522Artificial Analysis

Agents & tools

Instruction following

BenchmarkdefaultSource
IFBench28.1%#356Artificial Analysis

Long context

BenchmarkdefaultSource
AA-LCR28.0%#328Artificial Analysis

Composite

BenchmarkdefaultSource
AA Intelligence Index7.3#404Artificial Analysis

See also

Data as of 9 Sept 2026. Compare with another model.