Model comparison
GPT-5.4 nano vs Qwen3.5 122B-A10B
GPT-5.4 nano and Qwen3.5 122B-A10B score almost the same on the Noometry Index (41.9 vs 42.1), so choose on price, context window or the category you care about most.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. GPT-5.4 nano scores higher in 3 categories and Qwen3.5 122B-A10B in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where GPT-5.4 nano leads 43.6 to 39.1.
- The biggest single-benchmark swing is SciCode: 46.9% for GPT-5.4 nano and 35.6% for Qwen3.5 122B-A10B.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $0.40 / $3.20 for Qwen3.5 122B-A10B.
- GPT-5.4 nano accepts more context: 400K tokens versus 262K.
- Qwen3.5 122B-A10B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 nano | Qwen3.5 122B-A10B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.9 | 42.1 |
| Released | 2026-03-17 | 2026-02-23 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 66K |
| Input $ / M tokens | $0.20 | $0.40 |
| Output $ / M tokens | $1.25 | $3.20 |
| Results tracked | 40 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
GPT-5.4 nano: 43.6 (#84), Qwen3.5 122B-A10B: 39.1 (#162)
| Benchmark | GPT-5.4 nano | Qwen3.5 122B-A10B |
|---|---|---|
| SciCode | 46.9% | 35.6% |
| LMArena Coding | 1405 | 1436 |
| LMArena WebDev | — | 1360 |
| WeirdML | 49.2% | — |
| ALE-Bench | 1,005 | — |
Reasoning Qwen3.5 122B-A10B leads
GPT-5.4 nano: 23.7 (#173), Qwen3.5 122B-A10B: 27.2 (#123)
| Benchmark | GPT-5.4 nano | Qwen3.5 122B-A10B |
|---|---|---|
| CritPt | 9.3% | 0.9% |
| LMArena Hard Prompts | 1381 | 1421 |
| Mystery Game Puzzles | 9% | 17% |
| DTBench | 80.3% | 84.3% |
| LMCA | 36.9% | 32.2% |
| ARC-AGI-2 | 5.7% | — |
| Kagi LLM Benchmark | 39.7% | — |
| NYT Connections (extended) | — | 51.7% |
| ARC-AGI-1 | 51.5% | — |
| Chess Puzzles | 30% | — |
| Thematic Generalization | — | 51.2% |
| Epoch Capabilities Index | 145.81 | — |
| ForecastBench | 57.3 | — |
Math GPT-5.4 nano leads
GPT-5.4 nano: 40.9 (#88), Qwen3.5 122B-A10B: 39.1 (#112)
| Benchmark | GPT-5.4 nano | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Math | 1406 | 1432 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| OTIS Mock AIME 2024-2025 | 87.8% | — |
| ProofBench | 5% | — |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5.4 nano leads
GPT-5.4 nano: 41.9 (#103), Qwen3.5 122B-A10B: 38.8 (#142)
| Benchmark | GPT-5.4 nano | Qwen3.5 122B-A10B |
|---|---|---|
| Vectara Hallucination Rate | 3.1% | 11.2% |
| LMArena Expert | 1396 | 1432 |
| GPQA Diamond | 78.5% | — |
| SimpleQA Verified | 11.7% | — |
Multimodal Qwen3.5 122B-A10B leads
GPT-5.4 nano: 36.7 (#78), Qwen3.5 122B-A10B: 39.6 (#57)
| Benchmark | GPT-5.4 nano | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Vision | 1196 | 1245 |
Multilingual Qwen3.5 122B-A10B leads
GPT-5.4 nano: 48.6 (#140), Qwen3.5 122B-A10B: 51.6 (#107)
| Benchmark | GPT-5.4 nano | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Non-English | 1359 | 1400 |
| LMArena Chinese | 1392 | 1462 |
| LMArena French | 1396 | 1442 |
| LMArena German | 1367 | 1426 |
| LMArena Japanese | 1343 | 1367 |
| LMArena Korean | 1320 | 1352 |
| LMArena Russian | 1363 | 1400 |
| LMArena Spanish | 1371 | 1424 |
Instruction Following Qwen3.5 122B-A10B leads
GPT-5.4 nano: 71.9 (#144), Qwen3.5 122B-A10B: 73.8 (#115)
| Benchmark | GPT-5.4 nano | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Instruction Following | 1362 | 1399 |
Long Context Qwen3.5 122B-A10B leads
GPT-5.4 nano: 41.6 (#137), Qwen3.5 122B-A10B: 43.0 (#109)
| Benchmark | GPT-5.4 nano | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Longer Query | 1366 | 1410 |
Writing & Preference Qwen3.5 122B-A10B leads
GPT-5.4 nano: 55.7 (#142), Qwen3.5 122B-A10B: 60.0 (#105)
| Benchmark | GPT-5.4 nano | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Text | 1372 | 1417 |
| LMArena Creative Writing | 1314 | 1368 |
| LMArena Multi-Turn | 1382 | 1416 |
Frequently asked questions
Is GPT-5.4 nano better than Qwen3.5 122B-A10B?
GPT-5.4 nano and Qwen3.5 122B-A10B score almost the same on the Noometry Index (41.9 vs 42.1), so choose on price, context window or the category you care about most.
Which is cheaper, GPT-5.4 nano or Qwen3.5 122B-A10B?
GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; Qwen3.5 122B-A10B lists at $0.40 and $3.20.
Is GPT-5.4 nano or Qwen3.5 122B-A10B better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 39.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 nano does, with 400K tokens against 262K.
How many benchmarks do GPT-5.4 nano and Qwen3.5 122B-A10B share?
24 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Qwen3.5 122B-A10B has 27.