Noometry

Model comparison

Claude 3 Haiku vs Qwen3.8 27B

Qwen3.8 27B is the stronger model overall, scoring 46.0 to 25.9 on the Noometry Index.

Last verified . 22 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

Qwen3.8 27B Alibaba (Qwen)

46.0

Rank #68 Confirmed

Summary

  • They share 22 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Qwen3.8 27B in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Qwen3.8 27B leads 65.8 to 29.7.
  • The biggest single-benchmark swing is DTBench: 50.1% for Claude 3 Haiku and 88% for Qwen3.8 27B.
  • Qwen3.8 27B has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuQwen3.8 27B
ProviderAnthropicAlibaba (Qwen)
Noometry Index25.946.0
Released2024-03-072026-08-14
WeightsProprietaryOpen
Context window—262K
Max output—33K
Input $ / M tokens—$0.99
Output $ / M tokens—$1.49
Results tracked3731

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

Category by category

Coding Qwen3.8 27B leads

Claude 3 Haiku: 26.4 (#325), Qwen3.8 27B: 50.5 (#44)

BenchmarkClaude 3 HaikuQwen3.8 27B
LMArena Coding11991482
LMArena WebDev—1593
SciCode—46.6%
WeirdML9.8%—
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, Qwen3.8 27B: 32.9 (#57)

BenchmarkClaude 3 HaikuQwen3.8 27B
APEX-Agents—47.5%

Reasoning Qwen3.8 27B leads

Claude 3 Haiku: 16.3 (#307), Qwen3.8 27B: 41.0 (#54)

BenchmarkClaude 3 HaikuQwen3.8 27B
LMArena Hard Prompts11741460
DTBench50.1%88%
LMCA8.8%41.4%
Epoch Capabilities Index118.35149.38
ARC-AGI-2—42.4%
Kagi LLM Benchmark34.2%—
NYT Connections (extended)—54.5%
ARC-AGI-1—87.5%
CritPt—5.4%
Surface Evolver Bench—45%
ForecastBench53.2—
WinoGrande74.2%—

Math Qwen3.8 27B leads

Claude 3 Haiku: 9.8 (#319), Qwen3.8 27B: 37.1 (#161)

BenchmarkClaude 3 HaikuQwen3.8 27B
LMArena Math11881456
OTIS Mock AIME 2024-20251.8%—
ProofBench—16%
MATH Level 514.9%—

Knowledge Qwen3.8 27B leads

Claude 3 Haiku: 17.3 (#285), Qwen3.8 27B: 41.6 (#109)

BenchmarkClaude 3 HaikuQwen3.8 27B
LMArena Expert11481482
GPQA Diamond36.3%—
Confabulations34.2%—
MMLU73.8%—

Multimodal Qwen3.8 27B leads

Claude 3 Haiku: 23.6 (#128), Qwen3.8 27B: 41.3 (#37)

BenchmarkClaude 3 HaikuQwen3.8 27B
LMArena Vision9501271
ScienceQA72%—

Multilingual Qwen3.8 27B leads

Claude 3 Haiku: 36.0 (#243), Qwen3.8 27B: 53.7 (#60)

BenchmarkClaude 3 HaikuQwen3.8 27B
LMArena Non-English11781430
LMArena Chinese11551504
LMArena French11951465
LMArena German11741438
LMArena Japanese11021384
LMArena Korean11091393
LMArena Russian12041415
LMArena Spanish11661448

Instruction Following Qwen3.8 27B leads

Claude 3 Haiku: 61.3 (#247), Qwen3.8 27B: 75.8 (#53)

BenchmarkClaude 3 HaikuQwen3.8 27B
LMArena Instruction Following11731439

Long Context Qwen3.8 27B leads

Claude 3 Haiku: 36.1 (#237), Qwen3.8 27B: 44.3 (#70)

BenchmarkClaude 3 HaikuQwen3.8 27B
LMArena Longer Query11901450

Writing & Preference Qwen3.8 27B leads

Claude 3 Haiku: 29.7 (#291), Qwen3.8 27B: 65.8 (#43)

BenchmarkClaude 3 HaikuQwen3.8 27B
LMArena Text11951441
LMArena Creative Writing11571384
EQ-Bench Creative Writing7171671
LMArena Multi-Turn11901441

Frequently asked questions

Is Claude 3 Haiku better than Qwen3.8 27B?

Qwen3.8 27B is the stronger model overall, scoring 46.0 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or Qwen3.8 27B better for coding?

Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and Qwen3.8 27B share?

22 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Qwen3.8 27B has 31.

Related comparisons

Go deeper