Model comparison
GPT-6 Luna vs gpt-oss-20b
GPT-6 Luna is the stronger model overall, scoring 45.1 to 32.1 on the Noometry Index. gpt-oss-20b costs 5.6× less per token, which makes it the better buy when GPT-6 Luna's lead doesn't matter for your workload.
Last verified . 13 shared benchmarks.
Summary
- They share 13 benchmarks with published results for both. GPT-6 Luna scores higher in 7 categories and gpt-oss-20b in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-6 Luna leads 58.4 to 35.8.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.10 / $0.50 for GPT-6 Luna.
- GPT-6 Luna accepts more context: 1.05M tokens versus 131K.
- gpt-oss-20b has downloadable open weights; the other is API-only.
Side by side
| GPT-6 Luna | gpt-oss-20b | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 45.1 | 32.1 |
| Released | 2026-09-22 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 128K | 16K |
| Input $ / M tokens | $0.10 | $0.018 |
| Output $ / M tokens | $0.50 | $0.09 |
| Results tracked | 18 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
GPT-6 Luna: 54.7 (#24), gpt-oss-20b: 37.1 (#185)
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| LMArena Coding | 1448 | 1307 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| LMArena WebDev | 1583 | — |
| SciCode | — | 34.4% |
| WeirdML | — | 40.9% |
| ALE-Bench | — | 566.05 |
Agentic & Tool Use Not comparable
GPT-6 Luna: —, gpt-oss-20b: 9.2 (#146)
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
Reasoning GPT-6 Luna leads
GPT-6 Luna: 31.7 (#68), gpt-oss-20b: 18.8 (#251)
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| LMArena Hard Prompts | 1415 | 1274 |
| Kagi LLM Benchmark | — | 53.2% |
| NYT Connections (extended) | 68.7% | — |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 4% |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
| Epoch Capabilities Index | — | 137.81 |
Math gpt-oss-20b leads
GPT-6 Luna: 35.6 (#140), gpt-oss-20b: 37.6 (#93)
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| LMArena Math | 1396 | 1317 |
| OTIS Mock AIME 2024-2025 | — | 65.3% |
| Omni-MATH | — | 56.5% |
Knowledge GPT-6 Luna leads
GPT-6 Luna: 40.1 (#110), gpt-oss-20b: 34.6 (#186)
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| LMArena Expert | 1458 | 1258 |
| GPQA Diamond | — | 60.8% |
| MMLU-Pro | — | 74% |
| GPQA (HELM) | — | 59.4% |
Multimodal Not comparable
GPT-6 Luna: 42.0 (#27), gpt-oss-20b: —
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| Furniture Assembly | 44.2% | — |
Multilingual GPT-6 Luna leads
GPT-6 Luna: 51.0 (#111), gpt-oss-20b: 42.4 (#192)
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1390 | 1269 |
| LMArena Chinese | 1425 | 1311 |
| LMArena Russian | 1392 | 1279 |
| LMArena Spanish | 1416 | 1267 |
| LMArena German | — | 1249 |
| LMArena Japanese | — | 1245 |
| LMArena Korean | — | 1239 |
Instruction Following GPT-6 Luna leads
GPT-6 Luna: 74.7 (#85), gpt-oss-20b: 62.0 (#235)
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1414 | 1236 |
| IFEval | — | 73.2% |
Long Context GPT-6 Luna leads
GPT-6 Luna: 43.3 (#95), gpt-oss-20b: 37.9 (#203)
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1417 | 1251 |
Writing & Preference GPT-6 Luna leads
GPT-6 Luna: 58.4 (#115), gpt-oss-20b: 35.8 (#260)
| Benchmark | GPT-6 Luna | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1392 | 1288 |
| LMArena Creative Writing | 1357 | 1201 |
| LMArena Multi-Turn | 1400 | 1268 |
| EQ-Bench Creative Writing | — | 666 |
| WildBench | — | 73.7% |
Frequently asked questions
Is GPT-6 Luna better than gpt-oss-20b?
GPT-6 Luna is the stronger model overall, scoring 45.1 to 32.1 on the Noometry Index. gpt-oss-20b costs 5.6× less per token, which makes it the better buy when GPT-6 Luna's lead doesn't matter for your workload.
Which is cheaper, GPT-6 Luna or gpt-oss-20b?
gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; GPT-6 Luna lists at $0.10 and $0.50.
Is GPT-6 Luna or gpt-oss-20b better for coding?
GPT-6 Luna scores higher on coding benchmarks: 54.7 versus 37.1 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 131K.
How many benchmarks do GPT-6 Luna and gpt-oss-20b share?
13 benchmarks have published results for both models. GPT-6 Luna has 18 scored results on Noometry and gpt-oss-20b has 34.