Decision support

Kimi K3 vs X-Ai/Grok-4-Fast-Reasoning

Compare canonical objects directly. The useful dimensions change with the object type, so model and hardware comparisons stay separate.

Models Hardware

Kimi K3

Moonshot AI

Sources:models.devOpenRouter

X-Ai/Grok-4-Fast-Reasoning

xAI

Sources:models.dev

Benchmark results

Head to head on 2 benchmarks both models report

Kimi K32leads
X-Ai/Grok-4-Fast-Reasoning0leads

DeepSWE 1.1
Best pass@1 · Higher is better
- Kimi K3
  68.5%
  #8 of 28
- X-Ai/Grok-4-Fast-Reasoning
  Not reported
Runs and sources
Kimi K3
max68.5%mini-swe-agent · pass@4 89.4% · $4.65 / attempt
X-Ai/Grok-4-Fast-Reasoning
Not reported
Source:DeepSWE leaderboard, 2026-09-22 Official source JSON
Arena Text
Arena rating · Higher is better
- Kimi K3
  Best1487
  +82 ahead · Arena #19CI 1483 to 1491
- X-Ai/Grok-4-Fast-Reasoning
  1405
  Arena #161CI 1400 to 1410
Runs and sources
Kimi K3
max148795% CI 1483 to 1491 · 40,867 votes
X-Ai/Grok-4-Fast-Reasoning
default1405reasoning · 95% CI 1400 to 1410 · 18,206 votes
Source:Arena Text leaderboard, 2026-11-09
Arena WebDev
Arena rating · Higher is better
- Kimi K3
  Best1658
  +496 ahead · Arena #18CI 1652 to 1664
- X-Ai/Grok-4-Fast-Reasoning
  1162
  Arena #155CI 1134 to 1190
Runs and sources
Kimi K3
max165895% CI 1652 to 1664 · 22,405 votes
X-Ai/Grok-4-Fast-Reasoning
default1162reasoning · 95% CI 1134 to 1190 · 715 votes
Source:Arena WebDev leaderboard, 2026-11-10
Agent Arena
IPS score · Higher is better
- Kimi K3
  +0.006
  Arena #28CI -0.001 to +0.012
- X-Ai/Grok-4-Fast-Reasoning
  Not reported
Runs and sources
Kimi K3
max+0.00695% CI -0.001 to +0.012 · 16,124,377 observations
X-Ai/Grok-4-Fast-Reasoning
Not reported
Source:Agent Arena leaderboard, 2026-11-11
ARC-AGI-2
Best published score · Higher is better
- Kimi K3
  60.4%
  #36 of 85
- X-Ai/Grok-4-Fast-Reasoning
  Not reported
Runs and sources
Kimi K3
low12.4%
high55.0%
max60.4%
X-Ai/Grok-4-Fast-Reasoning
Not reported
Source:External evaluation via Epoch AIOriginal-source terms apply
FrontierMath-Tier-4-v2-Private
Best published score · Higher is better
- Kimi K3
  39.0%
  #28 of 72
- X-Ai/Grok-4-Fast-Reasoning
  Not reported
Runs and sources
Kimi K3
max39.0%
X-Ai/Grok-4-Fast-Reasoning
Not reported
Source:Epoch AI evaluation
FrontierMath-Tiers-1-3-v2-Private
Best published score · Higher is better
- Kimi K3
  72.2%
  #26 of 87
- X-Ai/Grok-4-Fast-Reasoning
  Not reported
Runs and sources
Kimi K3
max72.2%
X-Ai/Grok-4-Fast-Reasoning
Not reported
Source:Epoch AI evaluation
GPQA diamond
Best published score · Higher is better
- Kimi K3
  93.1%
  #21 of 161
- X-Ai/Grok-4-Fast-Reasoning
  Not reported
Runs and sources
Kimi K3
low84.8%
high91.9%
max93.1%
X-Ai/Grok-4-Fast-Reasoning
Not reported
Source:Epoch AI evaluation
OTIS Mock AIME 2024-2025
Best published score · Higher is better
- Kimi K3
  97.2%
  #32 of 158
- X-Ai/Grok-4-Fast-Reasoning
  Not reported
Runs and sources
Kimi K3
low68.9%
high93.3%
max97.2%
X-Ai/Grok-4-Fast-Reasoning
Not reported
Source:Epoch AI evaluation
SimpleQA Verified
Best published score · Higher is better
- Kimi K3
  50.6%
  #29 of 87
- X-Ai/Grok-4-Fast-Reasoning
  Not reported
Runs and sources
Kimi K3
max50.6%
X-Ai/Grok-4-Fast-Reasoning
Not reported
Source:Epoch AI evaluation

Each model's best published result is shown; "Not reported" means no matched evaluation, not a score of zero. Arena and DeepSWE ranks are the leaderboards' own; other ranks are among catalogue models with a score. Results from different harnesses or settings are not directly comparable.

Original-source terms apply to evaluations credited to their original publishers.

DeepSWE 1.1 results: official leaderboard, data generated 2026-09-22. Only exact model versions in the official dataset are matched.

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.

Model specifications

Release date

July 16, 2026

December 18, 2025

Context window

Native context

—

Max output tokens

943,718

2,000,000

Knowledge cutoff

—

Open weights

Closed

Parameter count

2.8T

—

API pricing · USD per 1M tokens

Input

$0.192

—

Output

$16.00

—

Cached input

$0.182

—

Cache write

—

Capabilities and modalities

Input modalities

text, image, video

text, image, audio, video

Output modalities

text

Tool calling

Supported

Reasoning option

Supported

Supported options

frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p

tools, reasoning, temperature

Capabilities

vision, thinking, completion, tools

—

Architecture and availability

Architecture

kimi-k3

—

Model families

kimi-k3

—

Tokenizer

Other

—

Instruction format

—

License

—

Providers

Provider endpoints

Ollama variants

Metadata source

ai-research

models.dev

A dash means the field is not reported. Provider-specific performance and pricing may differ.

Benchmark results

DeepSWE 1.1

Arena Text

Arena WebDev

Agent Arena

ARC-AGI-2

FrontierMath-Tier-4-v2-Private

FrontierMath-Tiers-1-3-v2-Private

GPQA diamond

OTIS Mock AIME 2024-2025

SimpleQA Verified

Model specifications

API pricing · USD per 1M tokens

Capabilities and modalities

Architecture and availability