Noometry

Model comparison

Gemini 2.5 Flash vs GPT-6 Astra

GPT-6 Astra is the stronger model overall, scoring 70.8 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 24× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.

Last verified . 32 shared benchmarks.

Gemini 2.5 Flash Google

39.3

Rank #170 Confirmed

GPT-6 Astra OpenAI

70.8

Rank #1 Confirmed

Summary

  • They share 32 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 1 category and GPT-6 Astra in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-6 Astra leads 85.1 to 18.1.
  • The biggest single-benchmark swing is ARC-AGI-2: 2.5% for Gemini 2.5 Flash and 95% for GPT-6 Astra.
  • Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $10 / $50 for GPT-6 Astra.
  • GPT-6 Astra accepts more context: 1.05M tokens versus 1.05M.

Side by side

Gemini 2.5 FlashGPT-6 Astra
ProviderGoogleOpenAI
Noometry Index39.370.8
Released2025-04-172026-09-03
WeightsProprietaryProprietary
Context window1.05M1.05M
Max output66K128K
Input $ / M tokens$0.30$10
Output $ / M tokens$2.50$50
Results tracked5456

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

Category by category

Coding GPT-6 Astra leads

Gemini 2.5 Flash: 35.8 (#220), GPT-6 Astra: 73.7 (#2)

BenchmarkGemini 2.5 FlashGPT-6 Astra
WeirdML41.9%93.6%
LMArena Coding14241487
ALE-Bench661.882,951
DeepSWE—74.1%
FrontierCode—53.3%
SWE-bench Verified (bash only)28.7%—
Aider Polyglot55.1%—
LMArena WebDev—1786
FrontierSWE—65.5%
SciCode—56.5%
GSO—79.4%
MirrorCode—46.7%

Agentic & Tool Use GPT-6 Astra leads

Gemini 2.5 Flash: 30.8 (#74), GPT-6 Astra: 52.9 (#3)

BenchmarkGemini 2.5 FlashGPT-6 Astra
BALROG33.5%68.3%
Vending-Bench 2548.8415,515
Terminal-Bench17.1%—
APEX-Agents—64.7%
Berkeley Function Calling Leaderboard56.2%—
Remote Labor Index—20.8%
TheAgentCompany41.1%—
GDP.pdf—34.2%

Reasoning GPT-6 Astra leads

Gemini 2.5 Flash: 18.1 (#286), GPT-6 Astra: 85.1 (#1)

BenchmarkGemini 2.5 FlashGPT-6 Astra
ARC-AGI-22.5%95%
ARC-AGI-133.3%98.5%
CritPt1.1%31.7%
LMArena Hard Prompts14221462
DTBench76.5%97.3%
LMCA27.5%64.4%
Epoch Capabilities Index143.03166.45
SimpleBench41.2%—
Kagi LLM Benchmark56.8%—
NYT Connections (extended)—98.1%
Chess Puzzles—72%
EnigmaEval2.7%—
EBR-Bench—76.2%
Mystery Game Puzzles—84%
Bench to the Future 3—0.14
ForecastBench60.6—

Math GPT-6 Astra leads

Gemini 2.5 Flash: 39.9 (#98), GPT-6 Astra: 93.5 (#2)

BenchmarkGemini 2.5 FlashGPT-6 Astra
OTIS Mock AIME 2024-202573.1%100%
LMArena Math14151465
FrontierMath (Tiers 1-3)—93.7%
FrontierMath Tier 4—97.6%
ProofBench—99%
Omni-MATH38.5%—
FrontierMath (Feb 2025 set)4.8%—
FrontierMath Erdős—2.9%
FrontierMath Tier 4 (v1)4.2%—

Knowledge GPT-6 Astra leads

Gemini 2.5 Flash: 36.4 (#168), GPT-6 Astra: 75.3 (#1)

BenchmarkGemini 2.5 FlashGPT-6 Astra
Humanity's Last Exam12.1%54.8%
Vectara Hallucination Rate7.8%8.7%
LMArena Expert14261483
GPQA Diamond—95.8%
SimpleQA Verified—75.6%
MMLU-Pro63.9%—
Confabulations16.8%—
GPQA (HELM)39%—

Multimodal GPT-6 Astra leads

Gemini 2.5 Flash: 41.8 (#32), GPT-6 Astra: 55.0 (#3)

BenchmarkGemini 2.5 FlashGPT-6 Astra
LMArena Vision12531281
GeoBench76%—
VPCT46.2%—
Blueprint-Bench 2—49.7%
Furniture Assembly—80%
LMArena Document—1468
SpatialViz-Bench36.9%—

Multilingual GPT-6 Astra leads

Gemini 2.5 Flash: 52.3 (#88), GPT-6 Astra: 53.7 (#61)

BenchmarkGemini 2.5 FlashGPT-6 Astra
LMArena Non-English14091430
LMArena Chinese14501484
LMArena French14331456
LMArena German14181440
LMArena Japanese14051379
LMArena Korean13851426
LMArena Russian14151436
LMArena Spanish14211407

Instruction Following Too close to call

Gemini 2.5 Flash: 75.7 (#54), GPT-6 Astra: 76.3 (#44)

BenchmarkGemini 2.5 FlashGPT-6 Astra
LMArena Instruction Following14051450
IFEval89.8%—

Long Context Gemini 2.5 Flash leads

Gemini 2.5 Flash: 47.5 (#17), GPT-6 Astra: 44.5 (#62)

BenchmarkGemini 2.5 FlashGPT-6 Astra
LMArena Longer Query14191456
Fiction.LiveBench77.8%—

Writing & Preference GPT-6 Astra leads

Gemini 2.5 Flash: 53.8 (#157), GPT-6 Astra: 75.3 (#7)

BenchmarkGemini 2.5 FlashGPT-6 Astra
LMArena Text14171441
LMArena Creative Writing14001418
EQ-Bench Creative Writing11372173
LMArena Multi-Turn14081448
Short-Story Creative Writing76.5%—
WildBench81.7%—

Frequently asked questions

Is Gemini 2.5 Flash better than GPT-6 Astra?

GPT-6 Astra is the stronger model overall, scoring 70.8 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 24× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.

Which is cheaper, Gemini 2.5 Flash or GPT-6 Astra?

Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; GPT-6 Astra lists at $10 and $50.

Is Gemini 2.5 Flash or GPT-6 Astra better for coding?

GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 35.8 in the Noometry coding category.

Which has the bigger context window?

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

How many benchmarks do Gemini 2.5 Flash and GPT-6 Astra share?

32 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and GPT-6 Astra has 56.

Related comparisons

Go deeper