Noometry

Model comparison

GPT-4 vs Qwen3.5 35B-A3B

Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 29.1 on the Noometry Index.

Last verified . 23 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Qwen3.5 35B-A3B Alibaba (Qwen)

42.0

Rank #123 Confirmed

Summary

  • They share 23 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Qwen3.5 35B-A3B in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Qwen3.5 35B-A3B leads 47.8 to 18.4.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 70% for Qwen3.5 35B-A3B.
  • Qwen3.5 35B-A3B is cheaper at $0.25 / $2 per million input/output tokens, against $30 / $60 for GPT-4.
  • Qwen3.5 35B-A3B accepts more context: 262K tokens versus 8K.
  • Qwen3.5 35B-A3B has downloadable open weights; the other is API-only.

Side by side

GPT-4Qwen3.5 35B-A3B
ProviderOpenAIAlibaba (Qwen)
Noometry Index29.142.0
Released2023-03-142026-02-01
WeightsProprietaryOpen
Context window8K262K
Max output8K66K
Input $ / M tokens$30$0.25
Output $ / M tokens$60$2
Results tracked3828

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

Category by category

Coding Qwen3.5 35B-A3B leads

GPT-4: 31.6 (#283), Qwen3.5 35B-A3B: 33.8 (#251)

BenchmarkGPT-4Qwen3.5 35B-A3B
LMArena Coding12541410
LMArena WebDev—1254
SciCode—29.3%
WeirdML12.4%—
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Qwen3.5 35B-A3B: —

BenchmarkGPT-4Qwen3.5 35B-A3B
METR Time Horizons36.1%—

Reasoning Qwen3.5 35B-A3B leads

GPT-4: 17.8 (#289), Qwen3.5 35B-A3B: 24.6 (#161)

BenchmarkGPT-4Qwen3.5 35B-A3B
Chess Puzzles4%10%
LMArena Hard Prompts12411400
DTBench62.7%80%
LMCA17.1%29.5%
Epoch Capabilities Index125.89142.52
CritPt—0.6%
Mystery Game Puzzles12%—
BIG-Bench Hard75.1%—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Qwen3.5 35B-A3B leads

GPT-4: 10.8 (#309), Qwen3.5 35B-A3B: 39.9 (#97)

BenchmarkGPT-4Qwen3.5 35B-A3B
OTIS Mock AIME 2024-20251.1%70%
LMArena Math12691404
MathArena Final-Answer Competitions—56%
MATH Level 523%—
GSM8K92%—

Knowledge Qwen3.5 35B-A3B leads

GPT-4: 18.4 (#282), Qwen3.5 35B-A3B: 47.8 (#79)

BenchmarkGPT-4Qwen3.5 35B-A3B
GPQA Diamond35.7%83.5%
LMArena Expert12111408
Vectara Hallucination Rate—10.5%
MMLU86.4%—
TriviaQA84.8%—

Multilingual Qwen3.5 35B-A3B leads

GPT-4: 40.6 (#215), Qwen3.5 35B-A3B: 50.0 (#127)

BenchmarkGPT-4Qwen3.5 35B-A3B
LMArena Non-English12461378
LMArena Chinese12421457
LMArena French12831412
LMArena German12511367
LMArena Japanese12091325
LMArena Korean11841356
LMArena Russian12511376
LMArena Spanish12611392

Instruction Following Qwen3.5 35B-A3B leads

GPT-4: 65.3 (#222), Qwen3.5 35B-A3B: 72.8 (#128)

BenchmarkGPT-4Qwen3.5 35B-A3B
LMArena Instruction Following12411379

Long Context Qwen3.5 35B-A3B leads

GPT-4: 37.7 (#212), Qwen3.5 35B-A3B: 42.4 (#127)

BenchmarkGPT-4Qwen3.5 35B-A3B
LMArena Longer Query12441389

Writing & Preference Qwen3.5 35B-A3B leads

GPT-4: 34.9 (#268), Qwen3.5 35B-A3B: 57.9 (#124)

BenchmarkGPT-4Qwen3.5 35B-A3B
LMArena Text12631395
LMArena Creative Writing12441346
LMArena Multi-Turn12571390
EQ-Bench Creative Writing752—

Frequently asked questions

Is GPT-4 better than Qwen3.5 35B-A3B?

Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 29.1 on the Noometry Index.

Which is cheaper, GPT-4 or Qwen3.5 35B-A3B?

Qwen3.5 35B-A3B is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or Qwen3.5 35B-A3B better for coding?

Qwen3.5 35B-A3B scores higher on coding benchmarks: 33.8 versus 31.6 in the Noometry coding category.

Which has the bigger context window?

Qwen3.5 35B-A3B does, with 262K tokens against 8K.

How many benchmarks do GPT-4 and Qwen3.5 35B-A3B share?

23 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Qwen3.5 35B-A3B has 28.

Related comparisons

Go deeper