Noometry

Model comparison

GPT-6 Astra vs Kimi K2 (Jul 2025)

GPT-6 Astra is the stronger model overall, scoring 70.8 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 20× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.

Last verified . 23 shared benchmarks.

GPT-6 Astra OpenAI

70.8

Rank #1 Confirmed

Kimi K2 (Jul 2025) Moonshot AI

41.2

Rank #140 Confirmed

Summary

  • They share 23 benchmarks with published results for both. GPT-6 Astra scores higher in 9 categories and Kimi K2 (Jul 2025) in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-6 Astra leads 85.1 to 23.3.
  • The biggest single-benchmark swing is GSO: 79.4% for GPT-6 Astra and 4.9% for Kimi K2 (Jul 2025).
  • Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $10 / $50 for GPT-6 Astra.
  • GPT-6 Astra accepts more context: 1.05M tokens versus 262K.
  • Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.

Side by side

GPT-6 AstraKimi K2 (Jul 2025)
ProviderOpenAIMoonshot AI
Noometry Index70.841.2
Released2026-09-032025-07-12
WeightsProprietaryOpen
Context window1.05M262K
Max output128K262K
Input $ / M tokens$10$0.57
Output $ / M tokens$50$2.30
Results tracked5642

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding GPT-6 Astra leads

GPT-6 Astra: 73.7 (#2), Kimi K2 (Jul 2025): 42.4 (#102)

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
GSO79.4%4.9%
WeirdML93.6%42.8%
LMArena Coding14871399
ALE-Bench2,951597.5
DeepSWE74.1%—
FrontierCode53.3%—
SWE-bench Verified (bash only)—63.4%
Aider Polyglot—59.1%
LMArena WebDev1786—
FrontierSWE65.5%—
SciCode56.5%—
MirrorCode46.7%—

Agentic & Tool Use GPT-6 Astra leads

GPT-6 Astra: 52.9 (#3), Kimi K2 (Jul 2025): 32.4 (#64)

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
Terminal-Bench—35.7%
APEX-Agents64.7%—
Berkeley Function Calling Leaderboard—59.1%
Remote Labor Index20.8%—
BALROG68.3%—
GDP.pdf34.2%—
METR Time Horizons—59.2%
Vending-Bench 215,515—

Reasoning GPT-6 Astra leads

GPT-6 Astra: 85.1 (#1), Kimi K2 (Jul 2025): 23.3 (#179)

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
LMArena Hard Prompts14621384
Epoch Capabilities Index166.45146.01
ARC-AGI-295%—
SimpleBench—26.3%
Kagi LLM Benchmark—64.4%
NYT Connections (extended)98.1%—
ARC-AGI-198.5%—
CritPt31.7%—
Chess Puzzles72%—
EBR-Bench76.2%—
Mystery Game Puzzles84%—
DTBench97.3%—
LMCA64.4%—
Bench to the Future 30.14—
ForecastBench—60.2

Math GPT-6 Astra leads

GPT-6 Astra: 93.5 (#2), Kimi K2 (Jul 2025): 42.7 (#83)

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
LMArena Math14651397
FrontierMath (Tiers 1-3)93.7%—
FrontierMath Tier 497.6%—
OTIS Mock AIME 2024-2025100%—
ProofBench99%—
Omni-MATH—65.4%
FrontierMath (Feb 2025 set)—21.4%
FrontierMath Erdős2.9%—
FrontierMath Tier 4 (v1)—0%

Knowledge GPT-6 Astra leads

GPT-6 Astra: 75.3 (#1), Kimi K2 (Jul 2025): 37.3 (#157)

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
Vectara Hallucination Rate8.7%17.9%
LMArena Expert14831365
GPQA Diamond95.8%—
Humanity's Last Exam54.8%—
SimpleQA Verified75.6%—
MMLU-Pro—81.9%
Confabulations—20.4%
GPQA (HELM)—65.3%

Multimodal Not comparable

GPT-6 Astra: 55.0 (#3), Kimi K2 (Jul 2025): —

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
LMArena Vision1281—
Blueprint-Bench 249.7%—
Furniture Assembly80%—
LMArena Document1468—

Multilingual GPT-6 Astra leads

GPT-6 Astra: 53.7 (#61), Kimi K2 (Jul 2025): 49.6 (#130)

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
LMArena Non-English14301372
LMArena Chinese14841415
LMArena French14561379
LMArena German14401387
LMArena Japanese13791349
LMArena Korean14261325
LMArena Russian14361385
LMArena Spanish14071386

Instruction Following GPT-6 Astra leads

GPT-6 Astra: 76.3 (#44), Kimi K2 (Jul 2025): 71.1 (#156)

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
LMArena Instruction Following14501348
IFEval—85%

Long Context GPT-6 Astra leads

GPT-6 Astra: 44.5 (#62), Kimi K2 (Jul 2025): 41.2 (#145)

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
LMArena Longer Query14561353
Fiction.LiveBench—66.7%
CL-bench—17.6%

Writing & Preference GPT-6 Astra leads

GPT-6 Astra: 75.3 (#7), Kimi K2 (Jul 2025): 62.3 (#78)

BenchmarkGPT-6 AstraKimi K2 (Jul 2025)
LMArena Text14411380
LMArena Creative Writing14181350
EQ-Bench Creative Writing21731666
LMArena Multi-Turn14481371
Short-Story Creative Writing—85.6%
WildBench—86.2%

Frequently asked questions

Is GPT-6 Astra better than Kimi K2 (Jul 2025)?

GPT-6 Astra is the stronger model overall, scoring 70.8 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 20× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.

Which is cheaper, GPT-6 Astra or Kimi K2 (Jul 2025)?

Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; GPT-6 Astra lists at $10 and $50.

Is GPT-6 Astra or Kimi K2 (Jul 2025) better for coding?

GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 42.4 in the Noometry coding category.

Which has the bigger context window?

GPT-6 Astra does, with 1.05M tokens against 262K.

How many benchmarks do GPT-6 Astra and Kimi K2 (Jul 2025) share?

23 benchmarks have published results for both models. GPT-6 Astra has 56 scored results on Noometry and Kimi K2 (Jul 2025) has 42.

Related comparisons

Go deeper