The Model Index
Up to four models

Qwen3-235B-A22B-Thinking (Jul 2025) vs Claude Haiku 4.5

On the index (GPT-4 = 100), Qwen3-235B-A22B-Thinking (Jul 2025) scores 138 and Claude Haiku 4.5 137. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Qwen3-235B-A22B-Thinking (Jul 2025)
Presets
Qwen3-235B-A22B-Thinking (Jul 2025)
138121–154
WeightsOpen
ReleasedJul 2025
Price in / out$0.45 / $3.5
Context128K
Core tests taken1 of 9
Claude Haiku 4.5
137121–153
WeightsClosed
ReleasedOct 2025
Price in / out$1.0 / $5.0
Context200K
Core tests taken3 of 9

Qwen3-235B-A22B-Thinking (Jul 2025) and Claude Haiku 4.5 are too close to call (1 point apart, within the uncertainty), ahead on 1 of 1 shared core tests, and costs 1.6× less.

Test by testthird-party results

Qwen3-235B-A22B-Thinking (Jul 2025)Claude Haiku 4.5
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)SimpleQA VerifiedSimpleQA VerifiedMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)FrontierMath (2025 set)FrontierMath (2025 set)FrontierMath tier 4 (2025 set)FrontierMath tier 4 (2025 set)CODINGWeirdMLWeirdMLAGENTSDTBenchDTBenchLMCALMCA
Every result as a table
TestQwen3-235B-A22B-Thinking (Jul 2025)Claude Haiku 4.5
Reasoning & knowledge
GPQA Diamond (Epoch AI)80.0%71.2%
SimpleQA Verified40.4%13.2%
Maths
AIME-style maths (OTIS mock)86.7%66.7%
FrontierMath (2025 set)8.5%5.9%
FrontierMath tier 4 (2025 set)0.0%2.1%
Coding
WeirdML41.0%45.4%
Agents
DTBench80.3%73.6%
LMCA29.3%30.9%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG