Model comparison
Claude Haiku 4.5 vs QwQ-32B
Claude Haiku 4.5 and QwQ-32B score almost the same on the Noometry Index (39.3 vs 39.4), so choose on price, context window or the category you care about most.
Last verified . 22 shared benchmarks.
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 5 categories and QwQ-32B in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude Haiku 4.5 leads 43.9 to 34.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.7% for Claude Haiku 4.5 and 59.2% for QwQ-32B.
- QwQ-32B has downloadable open weights; the other is API-only.
Side by side
| Claude Haiku 4.5 | QwQ-32B | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 39.3 | 39.4 |
| Released | 2025-10-15 | 2024-11-28 |
| Weights | Proprietary | Open |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $1 | — |
| Output $ / M tokens | $5 | — |
| Results tracked | 53 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.9 (#71), QwQ-32B: 34.9 (#220)
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| LMArena Coding | 1454 | 1333 |
| SWE-bench Verified (bash only) | 66.6% | — |
| Aider Polyglot | — | 20.9% |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| SciCode | 43.3% | — |
| WeirdML | 45.4% | — |
| BigCodeBench Instruct | — | 44.6% |
| LiveBench Coding | — | 72.2% |
| BigCodeBench Complete | — | 54.4% |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.5 (#50), QwQ-32B: —
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning QwQ-32B leads
Claude Haiku 4.5: 15.1 (#310), QwQ-32B: 22.9 (#171)
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| Chess Puzzles | 8% | 5% |
| LMArena Hard Prompts | 1420 | 1326 |
| Epoch Capabilities Index | 142.42 | 137.6 |
| ForecastBench | 61.4 | 58.3 |
| ARC-AGI-2 | 4% | — |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | — | 83.5% |
| DTBench | 73.6% | — |
| LiveBench Data Analysis | — | 65% |
| LMCA | 30.9% | — |
| LiveBench | — | 72% |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.0 (#69), QwQ-32B: 36.2 (#123)
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 59.2% |
| LMArena Math | 1395 | 1359 |
| Omni-MATH | 56.1% | — |
| LiveBench Math | — | 77.8% |
| MATH Level 5 | 96.4% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Too close to call
Claude Haiku 4.5: 37.6 (#146), QwQ-32B: 37.2 (#150)
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| GPQA Diamond | 71.2% | 65.3% |
| LMArena Expert | 1443 | 1323 |
| SimpleQA Verified | 13.2% | — |
| MMLU-Pro | 77.7% | — |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | 9.8% | — |
| GPQA (HELM) | 60.5% | — |
Multimodal Not comparable
Claude Haiku 4.5: 27.0 (#111), QwQ-32B: —
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Claude Haiku 4.5 leads
Claude Haiku 4.5: 50.1 (#124), QwQ-32B: 45.0 (#171)
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| LMArena Non-English | 1376 | 1306 |
| LMArena Chinese | 1415 | 1378 |
| LMArena French | 1408 | 1337 |
| LMArena German | 1377 | 1313 |
| LMArena Japanese | 1339 | 1263 |
| LMArena Korean | 1346 | 1278 |
| LMArena Russian | 1380 | 1297 |
| LMArena Spanish | 1420 | 1353 |
Instruction Following QwQ-32B leads
Claude Haiku 4.5: 71.6 (#144), QwQ-32B: 72.8 (#131)
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| LMArena Instruction Following | 1414 | 1297 |
| LiveBench Instruction Following | — | 81.8% |
| IFEval | 80.1% | — |
Long Context QwQ-32B leads
Claude Haiku 4.5: 43.6 (#86), QwQ-32B: 49.0 (#11)
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| LMArena Longer Query | 1427 | 1308 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 58.1 (#119), QwQ-32B: 50.9 (#175)
| Benchmark | Claude Haiku 4.5 | QwQ-32B |
|---|---|---|
| LMArena Text | 1396 | 1329 |
| LMArena Creative Writing | 1371 | 1288 |
| LMArena Multi-Turn | 1409 | 1314 |
| Short-Story Creative Writing | — | 80.2% |
| EQ-Bench Creative Writing | — | 1257 |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
| LiveBench Language | — | 51.4% |
Frequently asked questions
Is Claude Haiku 4.5 better than QwQ-32B?
Claude Haiku 4.5 and QwQ-32B score almost the same on the Noometry Index (39.3 vs 39.4), so choose on price, context window or the category you care about most.
Is Claude Haiku 4.5 or QwQ-32B better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 43.9 versus 34.9 in the Noometry coding category.
How many benchmarks do Claude Haiku 4.5 and QwQ-32B share?
22 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and QwQ-32B has 36.