Noometry

Model comparison

Claude Haiku 4.5 vs Qwen3-4B

Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 31.9 on the Noometry Index.

Last verified . 5 shared benchmarks.

Claude Haiku 4.5 Anthropic

39.5

Rank #165 Confirmed

Qwen3-4B Alibaba (Qwen)

31.9

Rank #264 Confirmed

Summary

  • They share 5 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 3 categories and Qwen3-4B in 1 category; 4 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Haiku 4.5 leads 44.9 to 29.7.
  • The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 68.7% for Claude Haiku 4.5 and 35.7% for Qwen3-4B.
  • Qwen3-4B has downloadable open weights; the other is API-only.

Side by side

Claude Haiku 4.5Qwen3-4B
ProviderAnthropicAlibaba (Qwen)
Noometry Index39.531.9
Released2025-10-152025-04-29
WeightsProprietaryOpen
Context window200K—
Max output64K—
Input $ / M tokens$1—
Output $ / M tokens$5—
Results tracked536

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

Category by category

Coding Not comparable

Claude Haiku 4.5: 44.0 (#78), Qwen3-4B: —

BenchmarkClaude Haiku 4.5Qwen3-4B
SWE-bench Verified (bash only)66.6%—
LMArena WebDev1330—
SWE-bench Multilingual64.7%—
SciCode43.3%—
WeirdML45.4%—
LMArena Coding1453—
ALE-Bench653.48—

Agentic & Tool Use Claude Haiku 4.5 leads

Claude Haiku 4.5: 33.6 (#52), Qwen3-4B: 27.6 (#100)

BenchmarkClaude Haiku 4.5Qwen3-4B
Berkeley Function Calling Leaderboard68.7%35.7%
Terminal-Bench35.5%—
DeepResearch Bench45.5%—
BALROG31.2%—
ExploitBench13.7%—
Vending-Bench 2458.89—

Reasoning Qwen3-4B leads

Claude Haiku 4.5: 15.1 (#320), Qwen3-4B: 19.2 (#268)

BenchmarkClaude Haiku 4.5Qwen3-4B
Chess Puzzles8%4%
ARC-AGI-24%—
NYT Connections (extended)14.3%—
ARC-AGI-147.7%—
CritPt0%—
LMArena Hard Prompts1420—
DTBench73.6%—
LMCA30.9%—
Epoch Capabilities Index142.41—
ForecastBench61.4—

Math Claude Haiku 4.5 leads

Claude Haiku 4.5: 44.9 (#78), Qwen3-4B: 29.7 (#240)

BenchmarkClaude Haiku 4.5Qwen3-4B
OTIS Mock AIME 2024-202566.7%52.2%
MathArena Final-Answer Competitions—38.5%
Omni-MATH56.1%—
LMArena Math1396—
MATH Level 596.4%—
FrontierMath (Feb 2025 set)5.9%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Claude Haiku 4.5 leads

Claude Haiku 4.5: 37.7 (#153), Qwen3-4B: 33.0 (#208)

BenchmarkClaude Haiku 4.5Qwen3-4B
GPQA Diamond71.2%52.3%
Vectara Hallucination Rate9.8%5.7%
SimpleQA Verified13.2%—
MMLU-Pro77.7%—
GPQA (HELM)60.5%—
LMArena Expert1442—

Multimodal Not comparable

Claude Haiku 4.5: 26.8 (#118), Qwen3-4B: —

BenchmarkClaude Haiku 4.5Qwen3-4B
Blueprint-Bench 20%—
LMArena Document1420—

Multilingual Not comparable

Claude Haiku 4.5: 49.9 (#129), Qwen3-4B: —

BenchmarkClaude Haiku 4.5Qwen3-4B
LMArena Non-English1377—
LMArena Chinese1417—
LMArena French1408—
LMArena German1375—
LMArena Japanese1339—
LMArena Korean1347—
LMArena Russian1381—
LMArena Spanish1420—

Instruction Following Not comparable

Claude Haiku 4.5: 71.4 (#149), Qwen3-4B: —

BenchmarkClaude Haiku 4.5Qwen3-4B
IFEval80.1%—
LMArena Instruction Following1414—

Long Context Not comparable

Claude Haiku 4.5: 43.6 (#92), Qwen3-4B: —

BenchmarkClaude Haiku 4.5Qwen3-4B
LMArena Longer Query1427—

Writing & Preference Not comparable

Claude Haiku 4.5: 57.9 (#123), Qwen3-4B: —

BenchmarkClaude Haiku 4.5Qwen3-4B
LMArena Text1396—
LMArena Creative Writing1372—
WildBench83.9%—
EQ-Bench 41064—
LMArena Multi-Turn1409—

Frequently asked questions

Is Claude Haiku 4.5 better than Qwen3-4B?

Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 31.9 on the Noometry Index.

How many benchmarks do Claude Haiku 4.5 and Qwen3-4B share?

5 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Qwen3-4B has 6.

Related comparisons

Go deeper