Noometry

Model comparison

GPT-4 Turbo vs Qwen3 8B

Qwen3 8B is the stronger model overall, scoring 33.7 to 30.5 on the Noometry Index.

Last verified . 6 shared benchmarks.

GPT-4 Turbo OpenAI

30.5

Rank #292 Confirmed

Qwen3 8B Alibaba (Qwen)

33.7

Rank #238 Confirmed

Summary

  • They share 6 benchmarks with published results for both. GPT-4 Turbo scores higher in 1 category and Qwen3 8B in 4 categories; 3 gaps are clear of the uncertainty.
  • The widest gap is in math, where Qwen3 8B leads 34.9 to 9.0.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.7% for GPT-4 Turbo and 56.1% for Qwen3 8B.
  • Qwen3 8B is cheaper at $0.18 / $0.70 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
  • Qwen3 8B accepts more context: 131K tokens versus 128K.
  • Qwen3 8B has downloadable open weights; the other is API-only.

Side by side

GPT-4 TurboQwen3 8B
ProviderOpenAIAlibaba (Qwen)
Noometry Index30.533.7
Released2023-11-062025-04
WeightsProprietaryOpen
Context window128K131K
Max output4K8K
Input $ / M tokens$10$0.18
Output $ / M tokens$30$0.70
Results tracked3611

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

Category by category

Coding Too close to call

GPT-4 Turbo: 33.8 (#249), Qwen3 8B: 34.0 (#248)

BenchmarkGPT-4 TurboQwen3 8B
SciCode—22.6%
WeirdML18%—
BigCodeBench Instruct48.2%—
LMArena Coding1268—
BigCodeBench Complete58.2%—
HumanEval+86.6%—
MBPP+73.3%—

Agentic & Tool Use Not comparable

GPT-4 Turbo: —, Qwen3 8B: 30.2 (#78)

BenchmarkGPT-4 TurboQwen3 8B
Berkeley Function Calling Leaderboard—42.6%
METR Time Horizons36.7%—

Reasoning Qwen3 8B leads

GPT-4 Turbo: 15.3 (#317), Qwen3 8B: 16.6 (#303)

BenchmarkGPT-4 TurboQwen3 8B
Chess Puzzles6%5%
DTBench61.6%59.7%
LMCA9.8%8.8%
Epoch Capabilities Index127.25136.17
SimpleBench25.1%—
CritPt—0%
LMArena Hard Prompts1251—
ForecastBench59.4—

Math Qwen3 8B leads

GPT-4 Turbo: 9.0 (#322), Qwen3 8B: 34.9 (#191)

BenchmarkGPT-4 TurboQwen3 8B
OTIS Mock AIME 2024-20256.7%56.1%
FrontierMath (Tiers 1-3)0.7%—
LMArena Math1272—
MATH Level 546.7%—

Knowledge Qwen3 8B leads

GPT-4 Turbo: 24.3 (#268), Qwen3 8B: 36.1 (#173)

BenchmarkGPT-4 TurboQwen3 8B
GPQA Diamond46.6%56.8%
Confabulations28.4%—
Vectara Hallucination Rate—4.8%
LMArena Expert1223—
MMLU81.3%—

Multimodal Not comparable

GPT-4 Turbo: 30.6 (#110), Qwen3 8B: —

BenchmarkGPT-4 TurboQwen3 8B
LMArena Vision1090—

Multilingual Not comparable

GPT-4 Turbo: 40.5 (#216), Qwen3 8B: —

BenchmarkGPT-4 TurboQwen3 8B
LMArena Non-English1245—
LMArena Chinese1242—
LMArena French1276—
LMArena German1259—
LMArena Japanese1194—
LMArena Korean1187—
LMArena Russian1259—
LMArena Spanish1260—

Instruction Following Not comparable

GPT-4 Turbo: 65.8 (#216), Qwen3 8B: —

BenchmarkGPT-4 TurboQwen3 8B
LMArena Instruction Following1249—

Long Context Too close to call

GPT-4 Turbo: 38.0 (#206), Qwen3 8B: 37.9 (#210)

BenchmarkGPT-4 TurboQwen3 8B
Fiction.LiveBench—62.1%
LMArena Longer Query1254—

Writing & Preference Not comparable

GPT-4 Turbo: 47.7 (#206), Qwen3 8B: —

BenchmarkGPT-4 TurboQwen3 8B
LMArena Text1272—
LMArena Creative Writing1269—
LMArena Multi-Turn1267—

Frequently asked questions

Is GPT-4 Turbo better than Qwen3 8B?

Qwen3 8B is the stronger model overall, scoring 33.7 to 30.5 on the Noometry Index.

Which is cheaper, GPT-4 Turbo or Qwen3 8B?

Qwen3 8B is cheaper. It lists at $0.18 per million input tokens and $0.70 per million output tokens; GPT-4 Turbo lists at $10 and $30.

Is GPT-4 Turbo or Qwen3 8B better for coding?

They score almost the same on coding (33.8 vs 34.0); test both on your own repository before choosing.

Which has the bigger context window?

Qwen3 8B does, with 131K tokens against 128K.

How many benchmarks do GPT-4 Turbo and Qwen3 8B share?

6 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Qwen3 8B has 11.

Related comparisons

Go deeper