Noometry

Model comparison

GPT-5 vs Qwen3-30B-A3B

GPT-5 is the stronger model overall, scoring 50.9 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 16× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.

Last verified . 30 shared benchmarks.

GPT-5 OpenAI

50.9

Rank #45 Confirmed

Qwen3-30B-A3B Alibaba (Qwen)

38.9

Rank #179 Confirmed

Summary

  • They share 30 benchmarks with published results for both. GPT-5 scores higher in 9 categories and Qwen3-30B-A3B in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in long context, where GPT-5 leads 69.5 to 31.0.
  • The biggest single-benchmark swing is Fiction.LiveBench: 97.2% for GPT-5 and 40.6% for Qwen3-30B-A3B.
  • Qwen3-30B-A3B is cheaper at $0.12 / $0.50 per million input/output tokens, against $1.25 / $10 for GPT-5.
  • GPT-5 accepts more context: 400K tokens versus 41K.
  • Qwen3-30B-A3B has downloadable open weights; the other is API-only.

Side by side

GPT-5Qwen3-30B-A3B
ProviderOpenAIAlibaba (Qwen)
Noometry Index50.938.9
Released2025-08-072025-04-28
WeightsProprietaryOpen
Context window400K41K
Max output128K16K
Input $ / M tokens$1.25$0.12
Output $ / M tokens$10$0.50
Results tracked6932

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

Category by category

Coding GPT-5 leads

GPT-5: 50.3 (#47), Qwen3-30B-A3B: 37.5 (#194)

BenchmarkGPT-5Qwen3-30B-A3B
SciCode42.9%33.3%
WeirdML60.7%29.8%
LMArena Coding14361416
SWE-bench Verified73.6%—
SWE-bench Verified (bash only)65%—
Aider Polyglot88%—
LMArena WebDev1418—
GSO6.9%—
ALE-Bench1,162—
AlgoTune1.67—

Agentic & Tool Use GPT-5 leads

GPT-5: 33.1 (#56), Qwen3-30B-A3B: 29.8 (#82)

BenchmarkGPT-5Qwen3-30B-A3B
Terminal-Bench49.6%—
Berkeley Function Calling Leaderboard—41.4%
GDPval34.8%—
Remote Labor Index1.7%—
DeepResearch Bench49.6%—
BALROG32.8%—
LMArena Search1133—
METR Time Horizons69.6%—

Reasoning GPT-5 leads

GPT-5: 38.3 (#64), Qwen3-30B-A3B: 22.2 (#204)

BenchmarkGPT-5Qwen3-30B-A3B
Kagi LLM Benchmark72.7%54.9%
CritPt12.6%0.3%
Chess Puzzles37%8%
LMArena Hard Prompts14161398
DTBench90.7%69.3%
LMCA40%22.4%
Epoch Capabilities Index150139.63
ARC-AGI-29.9%—
SimpleBench56.7%—
ARC-AGI-165.7%—
EnigmaEval10.5%—
EBR-Bench12.7%—
Mystery Game Puzzles23%—
ForecastBench61.4—

Math GPT-5 leads

GPT-5: 55.0 (#44), Qwen3-30B-A3B: 37.4 (#157)

BenchmarkGPT-5Qwen3-30B-A3B
OTIS Mock AIME 2024-202591.4%70.3%
LMArena Math14071394
FrontierMath (Tiers 1-3)55.4%—
FrontierMath Tier 422%—
MathArena Final-Answer Competitions—47.8%
ProofBench18%—
Omni-MATH64.7%—
MATH Level 598.1%—
FrontierMath (Feb 2025 set)32.4%—
FrontierMath Tier 4 (v1)12.5%—

Knowledge GPT-5 leads

GPT-5: 56.6 (#43), Qwen3-30B-A3B: 41.8 (#105)

BenchmarkGPT-5Qwen3-30B-A3B
GPQA Diamond86.2%70.1%
Confabulations10.3%12.3%
LMArena Expert14191396
Humanity's Last Exam25.3%—
SimpleQA Verified50.1%—
MMLU-Pro86.3%—
Vectara Hallucination Rate14.7%—
GPQA (HELM)79.2%—

Multimodal Not comparable

GPT-5: 46.8 (#13), Qwen3-30B-A3B: —

BenchmarkGPT-5Qwen3-30B-A3B
LMArena Vision1232—
GeoBench81%—
VPCT66%—

Multilingual GPT-5 leads

GPT-5: 51.4 (#110), Qwen3-30B-A3B: 49.5 (#132)

BenchmarkGPT-5Qwen3-30B-A3B
LMArena Non-English13971372
LMArena Chinese14221433
LMArena French14101418
LMArena German14161380
LMArena Japanese14091337
LMArena Korean13601331
LMArena Russian14061370
LMArena Spanish13991404

Instruction Following GPT-5 leads

GPT-5: 73.8 (#113), Qwen3-30B-A3B: 72.0 (#142)

BenchmarkGPT-5Qwen3-30B-A3B
LMArena Instruction Following13881363
IFEval87.5%—

Long Context GPT-5 leads

GPT-5: 69.5 (#2), Qwen3-30B-A3B: 31.0 (#283)

BenchmarkGPT-5Qwen3-30B-A3B
Fiction.LiveBench97.2%40.6%
LMArena Longer Query13991379

Writing & Preference GPT-5 leads

GPT-5: 63.4 (#65), Qwen3-30B-A3B: 55.6 (#143)

BenchmarkGPT-5Qwen3-30B-A3B
LMArena Text14061384
LMArena Creative Writing13651317
Short-Story Creative Writing86%75.3%
LMArena Multi-Turn14261378
EQ-Bench Creative Writing1627—
WildBench85.7%—

Frequently asked questions

Is GPT-5 better than Qwen3-30B-A3B?

GPT-5 is the stronger model overall, scoring 50.9 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 16× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.

Which is cheaper, GPT-5 or Qwen3-30B-A3B?

Qwen3-30B-A3B is cheaper. It lists at $0.12 per million input tokens and $0.50 per million output tokens; GPT-5 lists at $1.25 and $10.

Is GPT-5 or Qwen3-30B-A3B better for coding?

GPT-5 scores higher on coding benchmarks: 50.3 versus 37.5 in the Noometry coding category.

Which has the bigger context window?

GPT-5 does, with 400K tokens against 41K.

How many benchmarks do GPT-5 and Qwen3-30B-A3B share?

30 benchmarks have published results for both models. GPT-5 has 69 scored results on Noometry and Qwen3-30B-A3B has 32.

Related comparisons

Go deeper