Model comparison
gpt-oss-20b vs Tulu 3 (Tülu 3) 70B
gpt-oss-20b and Tulu 3 (Tülu 3) 70B score almost the same on the Noometry Index (32.5 vs 33.0), so choose on price, context window or the category you care about most.
Last verified . 13 shared benchmarks.
Allen Institute for AI (Ai2)Summary
- They share 13 benchmarks with published results for both. gpt-oss-20b scores higher in 5 categories and Tulu 3 (Tülu 3) 70B in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-20b leads 39.4 to 14.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 65.3% for gpt-oss-20b and 4.4% for Tulu 3 (Tülu 3) 70B.
Side by side
| gpt-oss-20b | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | OpenAI | Allen Institute for AI (Ai2) |
| Noometry Index | 32.5 | 33.0 |
| Released | 2025-08-05 | 2024-11-21 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.018 | — |
| Output $ / M tokens | $0.09 | — |
| Results tracked | 34 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-20b leads
gpt-oss-20b: 37.6 (#192), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | gpt-oss-20b | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1306 | 1235 |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
| ALE-Bench | 566.05 | — |
Agentic & Tool Use Not comparable
gpt-oss-20b: 9.3 (#154), Tulu 3 (Tülu 3) 70B: —
| Benchmark | gpt-oss-20b | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| Terminal-Bench | 3.4% | — |
Reasoning Tulu 3 (Tülu 3) 70B leads
gpt-oss-20b: 19.3 (#261), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | gpt-oss-20b | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1274 | 1220 |
| Kagi LLM Benchmark | 53.2% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
| Epoch Capabilities Index | 137.82 | — |
Math gpt-oss-20b leads
gpt-oss-20b: 39.4 (#103), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | gpt-oss-20b | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3% | 4.4% |
| LMArena Math | 1317 | 1242 |
| Omni-MATH | 56.5% | — |
| MATH Level 5 | — | 42.7% |
Knowledge gpt-oss-20b leads
gpt-oss-20b: 34.6 (#195), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | gpt-oss-20b | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | 60.8% | 46.3% |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |
| LMArena Expert | 1258 | — |
Multilingual gpt-oss-20b leads
gpt-oss-20b: 42.2 (#197), Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | gpt-oss-20b | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1268 | 1236 |
| LMArena Chinese | 1314 | 1249 |
| LMArena Russian | 1278 | 1246 |
| LMArena German | 1255 | — |
| LMArena Japanese | 1244 | — |
| LMArena Korean | 1236 | — |
| LMArena Spanish | 1267 | — |
Instruction Following Tulu 3 (Tülu 3) 70B leads
gpt-oss-20b: 61.8 (#240), Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | gpt-oss-20b | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1236 | 1233 |
| IFEval | 73.2% | — |
Long Context Too close to call
gpt-oss-20b: 37.9 (#209), Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | gpt-oss-20b | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1250 | 1224 |
Writing & Preference Tulu 3 (Tülu 3) 70B leads
gpt-oss-20b: 35.5 (#265), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | gpt-oss-20b | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1287 | 1256 |
| LMArena Creative Writing | 1201 | 1231 |
| LMArena Multi-Turn | 1268 | 1252 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
Frequently asked questions
Is gpt-oss-20b better than Tulu 3 (Tülu 3) 70B?
gpt-oss-20b and Tulu 3 (Tülu 3) 70B score almost the same on the Noometry Index (32.5 vs 33.0), so choose on price, context window or the category you care about most.
Is gpt-oss-20b or Tulu 3 (Tülu 3) 70B better for coding?
gpt-oss-20b scores higher on coding benchmarks: 37.6 versus 36.0 in the Noometry coding category.
How many benchmarks do gpt-oss-20b and Tulu 3 (Tülu 3) 70B share?
13 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.