Model comparison
GPT-5.6 Luna vs gpt-oss-120b
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 36.3 on the Noometry Index. gpt-oss-120b costs 6.4× less per token, which makes it the better buy when GPT-5.6 Luna's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. GPT-5.6 Luna scores higher in 9 categories and gpt-oss-120b in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.6 Luna leads 47.6 to 20.0.
- The biggest single-benchmark swing is APEX-Agents: 43% for GPT-5.6 Luna and 4.4% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.20 / $1.20 for GPT-5.6 Luna.
- GPT-5.6 Luna accepts more context: 1.05M tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.
Side by side
| GPT-5.6 Luna | gpt-oss-120b | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 54.6 | 36.3 |
| Released | 2026-07-09 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 128K | 41K |
| Input $ / M tokens | $0.20 | $0.037 |
| Output $ / M tokens | $1.20 | $0.17 |
| Results tracked | 52 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
GPT-5.6 Luna: 54.5 (#28), gpt-oss-120b: 33.5 (#256)
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| SciCode | 53.6% | 36% |
| WeirdML | 60.9% | 48.2% |
| LMArena Coding | 1466 | 1380 |
| ALE-Bench | 1,667 | 575.62 |
| DeepSWE | 67.2% | — |
| FrontierCode | 39.8% | — |
| SWE-bench Verified (bash only) | — | 26% |
| Aider Polyglot | — | 41.8% |
| CursorBench | 35.9% | — |
| LMArena WebDev | 1519 | — |
| AlgoTune | — | 1.41 |
Agentic & Tool Use GPT-5.6 Luna leads
GPT-5.6 Luna: 34.4 (#45), gpt-oss-120b: 12.2 (#153)
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| APEX-Agents | 43% | 4.4% |
| Vending-Bench 2 | 4,095 | -21.53 |
| Terminal-Bench | — | 18.7% |
| BALROG | 45.6% | — |
| GDP.pdf | 22.7% | — |
| METR Time Horizons | — | 56.6% |
Reasoning GPT-5.6 Luna leads
GPT-5.6 Luna: 47.6 (#43), gpt-oss-120b: 20.0 (#245)
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| SimpleBench | 46.8% | 22.1% |
| Kagi LLM Benchmark | 49.1% | 58.6% |
| CritPt | 20.6% | 1.1% |
| Chess Puzzles | 40% | 20% |
| LMArena Hard Prompts | 1451 | 1364 |
| Mystery Game Puzzles | 21% | 2% |
| DTBench | 89.1% | 76.3% |
| LMCA | 48.5% | 22.1% |
| Surface Evolver Bench | 61.9% | 25% |
| Epoch Capabilities Index | 156.39 | 139.93 |
| ARC-AGI-2 | 59.5% | — |
| NYT Connections (extended) | 69.4% | — |
| ARC-AGI-1 | 88% | — |
Math GPT-5.6 Luna leads
GPT-5.6 Luna: 77.7 (#14), gpt-oss-120b: 52.5 (#50)
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 88.9% |
| LMArena Math | 1458 | 1389 |
| FrontierMath (Tiers 1-3) | 82.1% | — |
| FrontierMath Tier 4 | 61% | — |
| ProofBench | 60% | — |
| Omni-MATH | — | 68.8% |
Knowledge GPT-5.6 Luna leads
GPT-5.6 Luna: 58.5 (#34), gpt-oss-120b: 42.4 (#96)
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 91.6% | 75.8% |
| LMArena Expert | 1478 | 1356 |
| SimpleQA Verified | 41% | — |
| MMLU-Pro | — | 79.5% |
| Confabulations | — | 15.7% |
| Vectara Hallucination Rate | — | 14.2% |
| GPQA (HELM) | — | 68.4% |
Multimodal Not comparable
GPT-5.6 Luna: 42.7 (#28), gpt-oss-120b: —
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| LMArena Vision | 1258 | — |
| Blueprint-Bench 2 | 22.6% | — |
| Furniture Assembly | 42.5% | — |
| LMArena Document | 1457 | — |
Multilingual GPT-5.6 Luna leads
GPT-5.6 Luna: 52.8 (#78), gpt-oss-120b: 48.0 (#147)
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1417 | 1351 |
| LMArena Chinese | 1470 | 1385 |
| LMArena French | 1456 | 1369 |
| LMArena German | 1454 | 1353 |
| LMArena Japanese | 1411 | 1331 |
| LMArena Korean | 1415 | 1282 |
| LMArena Russian | 1428 | 1343 |
| LMArena Spanish | 1448 | 1389 |
Instruction Following GPT-5.6 Luna leads
GPT-5.6 Luna: 75.6 (#57), gpt-oss-120b: 69.3 (#173)
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| LMArena Instruction Following | 1437 | 1318 |
| IFEval | — | 83.6% |
Long Context GPT-5.6 Luna leads
GPT-5.6 Luna: 43.9 (#82), gpt-oss-120b: 31.4 (#278)
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| LMArena Longer Query | 1436 | 1319 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference GPT-5.6 Luna leads
GPT-5.6 Luna: 68.0 (#29), gpt-oss-120b: 46.5 (#217)
| Benchmark | GPT-5.6 Luna | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1431 | 1365 |
| LMArena Creative Writing | 1396 | 1275 |
| EQ-Bench Creative Writing | 1829 | 961 |
| LMArena Multi-Turn | 1434 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
| WildBench | — | 84.5% |
| EQ-Bench 4 | 1156 | — |
Frequently asked questions
Is GPT-5.6 Luna better than gpt-oss-120b?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 36.3 on the Noometry Index. gpt-oss-120b costs 6.4× less per token, which makes it the better buy when GPT-5.6 Luna's lead doesn't matter for your workload.
Which is cheaper, GPT-5.6 Luna 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-5.6 Luna lists at $0.20 and $1.20.
Is GPT-5.6 Luna or gpt-oss-120b better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Luna does, with 1.05M tokens against 131K.
How many benchmarks do GPT-5.6 Luna and gpt-oss-120b share?
34 benchmarks have published results for both models. GPT-5.6 Luna has 52 scored results on Noometry and gpt-oss-120b has 48.