Model comparison
GPT-5.4 nano vs gpt-oss-120b
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 36.3 on the Noometry Index. gpt-oss-120b costs 6.6× less per token, which makes it the better buy when GPT-5.4 nano's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. GPT-5.4 nano scores higher in 6 categories and gpt-oss-120b in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 52.5 to 40.9.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 39.7% for GPT-5.4 nano and 58.6% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.20 / $1.25 for GPT-5.4 nano.
- GPT-5.4 nano accepts more context: 400K tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 nano | gpt-oss-120b | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 41.9 | 36.3 |
| Released | 2026-03-17 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 400K | 131K |
| Max output | 128K | 41K |
| Input $ / M tokens | $0.20 | $0.037 |
| Output $ / M tokens | $1.25 | $0.17 |
| Results tracked | 40 | 48 |
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), gpt-oss-120b: 33.5 (#256)
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| SciCode | 46.9% | 36% |
| WeirdML | 49.2% | 48.2% |
| LMArena Coding | 1405 | 1380 |
| ALE-Bench | 1,005 | 575.62 |
| SWE-bench Verified (bash only) | — | 26% |
| Aider Polyglot | — | 41.8% |
| AlgoTune | — | 1.41 |
Agentic & Tool Use Not comparable
GPT-5.4 nano: —, gpt-oss-120b: 12.2 (#153)
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| Terminal-Bench | — | 18.7% |
| APEX-Agents | — | 4.4% |
| METR Time Horizons | — | 56.6% |
| Vending-Bench 2 | — | -21.53 |
Reasoning GPT-5.4 nano leads
GPT-5.4 nano: 23.7 (#173), gpt-oss-120b: 20.0 (#245)
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| Kagi LLM Benchmark | 39.7% | 58.6% |
| CritPt | 9.3% | 1.1% |
| Chess Puzzles | 30% | 20% |
| LMArena Hard Prompts | 1381 | 1364 |
| Mystery Game Puzzles | 9% | 2% |
| DTBench | 80.3% | 76.3% |
| LMCA | 36.9% | 22.1% |
| Epoch Capabilities Index | 145.81 | 139.93 |
| ARC-AGI-2 | 5.7% | — |
| SimpleBench | — | 22.1% |
| ARC-AGI-1 | 51.5% | — |
| Surface Evolver Bench | — | 25% |
| ForecastBench | 57.3 | — |
Math gpt-oss-120b leads
GPT-5.4 nano: 40.9 (#88), gpt-oss-120b: 52.5 (#50)
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 88.9% |
| LMArena Math | 1406 | 1389 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 5% | — |
| Omni-MATH | — | 68.8% |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge Too close to call
GPT-5.4 nano: 41.9 (#103), gpt-oss-120b: 42.4 (#96)
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 78.5% | 75.8% |
| Vectara Hallucination Rate | 3.1% | 14.2% |
| LMArena Expert | 1396 | 1356 |
| SimpleQA Verified | 11.7% | — |
| MMLU-Pro | — | 79.5% |
| Confabulations | — | 15.7% |
| GPQA (HELM) | — | 68.4% |
Multimodal Not comparable
GPT-5.4 nano: 36.7 (#78), gpt-oss-120b: —
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| LMArena Vision | 1196 | — |
Multilingual Too close to call
GPT-5.4 nano: 48.6 (#140), gpt-oss-120b: 48.0 (#147)
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1359 | 1351 |
| LMArena Chinese | 1392 | 1385 |
| LMArena French | 1396 | 1369 |
| LMArena German | 1367 | 1353 |
| LMArena Japanese | 1343 | 1331 |
| LMArena Korean | 1320 | 1282 |
| LMArena Russian | 1363 | 1343 |
| LMArena Spanish | 1371 | 1389 |
Instruction Following GPT-5.4 nano leads
GPT-5.4 nano: 71.9 (#144), gpt-oss-120b: 69.3 (#173)
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| LMArena Instruction Following | 1362 | 1318 |
| IFEval | — | 83.6% |
Long Context GPT-5.4 nano leads
GPT-5.4 nano: 41.6 (#137), gpt-oss-120b: 31.4 (#278)
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| LMArena Longer Query | 1366 | 1319 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference GPT-5.4 nano leads
GPT-5.4 nano: 55.7 (#142), gpt-oss-120b: 46.5 (#217)
| Benchmark | GPT-5.4 nano | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1372 | 1365 |
| LMArena Creative Writing | 1314 | 1275 |
| LMArena Multi-Turn | 1382 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
| EQ-Bench Creative Writing | — | 961 |
| WildBench | — | 84.5% |
Frequently asked questions
Is GPT-5.4 nano better than gpt-oss-120b?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 36.3 on the Noometry Index. gpt-oss-120b costs 6.6× less per token, which makes it the better buy when GPT-5.4 nano's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 nano or gpt-oss-120b?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; GPT-5.4 nano lists at $0.20 and $1.25.
Is GPT-5.4 nano or gpt-oss-120b better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 nano does, with 400K tokens against 131K.
How many benchmarks do GPT-5.4 nano and gpt-oss-120b share?
30 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and gpt-oss-120b has 48.