Model comparison
o4-mini vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 41.6 on the Noometry Index. o4-mini costs 1.6× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
Alibaba (Qwen)Summary
- They share 29 benchmarks with published results for both. o4-mini scores higher in 1 category and Qwen3.8 Max in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 40.8.
- The biggest single-benchmark swing is FrontierMath Tier 4: 4.9% for o4-mini and 46.3% for Qwen3.8 Max.
- o4-mini is cheaper at $1.10 / $4.40 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 200K.
Side by side
| o4-mini | Qwen3.8 Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.6 | 56.8 |
| Released | 2025-04-16 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1M |
| Max output | 100K | 131K |
| Input $ / M tokens | $1.10 | $2 |
| Output $ / M tokens | $4.40 | $6 |
| Results tracked | 60 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
o4-mini: 40.9 (#127), Qwen3.8 Max: 53.5 (#29)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| LMArena Coding | 1368 | 1502 |
| DeepSWE | — | 57.5% |
| SWE-bench Verified (bash only) | 45% | — |
| Aider Polyglot | 72% | — |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| GSO | 3.6% | — |
| WeirdML | 52.6% | — |
| CadEval | 62% | — |
| ALE-Bench | 826.17 | — |
| AlgoTune | 1.72 | — |
Agentic & Tool Use Qwen3.8 Max leads
o4-mini: 32.6 (#61), Qwen3.8 Max: 45.4 (#14)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| Berkeley Function Calling Leaderboard | 53.2% | — |
| GDPval | 25.3% | — |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
| METR Time Horizons | 63.9% | — |
Reasoning Qwen3.8 Max leads
o4-mini: 24.6 (#162), Qwen3.8 Max: 54.4 (#26)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| CritPt | 0.6% | 20% |
| Chess Puzzles | 26% | 40% |
| LMArena Hard Prompts | 1351 | 1496 |
| Mystery Game Puzzles | 5% | 38% |
| DTBench | 77.6% | 92% |
| LMCA | 26.5% | 46.2% |
| Epoch Capabilities Index | 145.64 | 156.41 |
| ARC-AGI-2 | 6.1% | — |
| SimpleBench | 38.7% | — |
| Kagi LLM Benchmark | 67.6% | — |
| NYT Connections (extended) | — | 88.3% |
| ARC-AGI-1 | 58.7% | — |
| EnigmaEval | 9.2% | — |
| ForecastBench | 61.8 | — |
Math Qwen3.8 Max leads
o4-mini: 40.8 (#89), Qwen3.8 Max: 73.2 (#20)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 36.1% | 74.7% |
| FrontierMath Tier 4 | 4.9% | 46.3% |
| OTIS Mock AIME 2024-2025 | 81.7% | 100% |
| LMArena Math | 1389 | 1499 |
| ProofBench | — | 58% |
| Omni-MATH | 72% | — |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 24.8% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge Qwen3.8 Max leads
o4-mini: 43.6 (#91), Qwen3.8 Max: 61.7 (#27)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 79.6% | 92.7% |
| SimpleQA Verified | 19.6% | 47.3% |
| LMArena Expert | 1343 | 1507 |
| Humanity's Last Exam | 18.1% | — |
| MMLU-Pro | 82% | — |
| Confabulations | 15.8% | — |
| Vectara Hallucination Rate | 18.6% | — |
| GPQA (HELM) | 73.5% | — |
Multimodal o4-mini leads
o4-mini: 40.2 (#49), Qwen3.8 Max: 37.2 (#75)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 1194 | 1314 |
| GeoBench | 64% | — |
| VPCT | 57.5% | — |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
o4-mini: 47.0 (#154), Qwen3.8 Max: 56.7 (#18)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1337 | 1472 |
| LMArena Chinese | 1354 | 1538 |
| LMArena French | 1364 | 1503 |
| LMArena German | 1336 | 1483 |
| LMArena Japanese | 1308 | 1467 |
| LMArena Korean | 1312 | 1461 |
| LMArena Russian | 1334 | 1481 |
| LMArena Spanish | 1347 | 1492 |
Instruction Following Qwen3.8 Max leads
o4-mini: 75.2 (#68), Qwen3.8 Max: 77.6 (#17)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1321 | 1479 |
| IFEval | 92.8% | — |
Long Context Too close to call
o4-mini: 45.5 (#33), Qwen3.8 Max: 45.6 (#31)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1315 | 1489 |
| Fiction.LiveBench | 77.8% | — |
Writing & Preference Qwen3.8 Max leads
o4-mini: 54.0 (#152), Qwen3.8 Max: 67.1 (#30)
| Benchmark | o4-mini | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1353 | 1483 |
| LMArena Creative Writing | 1294 | 1479 |
| LMArena Multi-Turn | 1350 | 1489 |
| Short-Story Creative Writing | 75% | — |
| WildBench | 85.4% | — |
Frequently asked questions
Is o4-mini better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 41.6 on the Noometry Index. o4-mini costs 1.6× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, o4-mini or Qwen3.8 Max?
o4-mini is cheaper. It lists at $1.10 per million input tokens and $4.40 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is o4-mini or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 40.9 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 Max does, with 1M tokens against 200K.
How many benchmarks do o4-mini and Qwen3.8 Max share?
29 benchmarks have published results for both models. o4-mini has 60 scored results on Noometry and Qwen3.8 Max has 39.