Model comparison
Claude Haiku 4.5 vs Qwen2.5-Max
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 39.5 on the Noometry Index.
Last verified . 18 shared benchmarks.
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 7 categories and Qwen2.5-Max in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen2.5-Max leads 25.6 to 15.1.
Side by side
| Claude Haiku 4.5 | Qwen2.5-Max | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 39.5 | 40.7 |
| Released | 2025-10-15 | 2025-01-25 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $1 | — |
| Output $ / M tokens | $5 | — |
| Results tracked | 53 | 27 |
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: 44.0 (#78), Qwen2.5-Max: 41.8 (#117)
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1453 | 1359 |
| SWE-bench Verified (bash only) | 66.6% | — |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| SciCode | 43.3% | — |
| WeirdML | 45.4% | — |
| LiveBench Coding | — | 64.4% |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.6 (#52), Qwen2.5-Max: —
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| 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 Qwen2.5-Max leads
Claude Haiku 4.5: 15.1 (#320), Qwen2.5-Max: 25.6 (#147)
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1420 | 1360 |
| Epoch Capabilities Index | 142.41 | 132.53 |
| ARC-AGI-2 | 4% | — |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| CritPt | 0% | — |
| Chess Puzzles | 8% | — |
| LiveBench Reasoning | — | 51.4% |
| DTBench | 73.6% | — |
| LiveBench Data Analysis | — | 67.9% |
| LMCA | 30.9% | — |
| ForecastBench | 61.4 | — |
| LiveBench | — | 62.3% |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 44.9 (#78), Qwen2.5-Max: 36.9 (#162)
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1396 | 1369 |
| OTIS Mock AIME 2024-2025 | 66.7% | — |
| Omni-MATH | 56.1% | — |
| LiveBench Math | — | 58.4% |
| MATH Level 5 | 96.4% | — |
| 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), Qwen2.5-Max: 35.3 (#186)
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| LMArena Expert | 1442 | 1337 |
| GPQA Diamond | 71.2% | — |
| SimpleQA Verified | 13.2% | — |
| MMLU-Pro | 77.7% | — |
| Confabulations | — | 21.8% |
| Vectara Hallucination Rate | 9.8% | — |
| GPQA (HELM) | 60.5% | — |
Multimodal Not comparable
Claude Haiku 4.5: 26.8 (#118), Qwen2.5-Max: —
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Claude Haiku 4.5 leads
Claude Haiku 4.5: 49.9 (#129), Qwen2.5-Max: 48.1 (#146)
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1377 | 1352 |
| LMArena Chinese | 1417 | 1382 |
| LMArena French | 1408 | 1396 |
| LMArena German | 1375 | 1350 |
| LMArena Japanese | 1339 | 1300 |
| LMArena Korean | 1347 | 1304 |
| LMArena Russian | 1381 | 1353 |
| LMArena Spanish | 1420 | 1377 |
Instruction Following Too close to call
Claude Haiku 4.5: 71.4 (#149), Qwen2.5-Max: 71.3 (#152)
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1414 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
| IFEval | 80.1% | — |
Long Context Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.6 (#92), Qwen2.5-Max: 41.4 (#142)
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1427 | 1358 |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 57.9 (#123), Qwen2.5-Max: 55.4 (#146)
| Benchmark | Claude Haiku 4.5 | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1396 | 1367 |
| LMArena Creative Writing | 1372 | 1339 |
| LMArena Multi-Turn | 1409 | 1364 |
| Short-Story Creative Writing | — | 72.9% |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is Claude Haiku 4.5 better than Qwen2.5-Max?
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 39.5 on the Noometry Index.
Is Claude Haiku 4.5 or Qwen2.5-Max better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 44.0 versus 41.8 in the Noometry coding category.
How many benchmarks do Claude Haiku 4.5 and Qwen2.5-Max share?
18 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Qwen2.5-Max has 27.