Model comparison
gpt-oss-20b vs Llama 3.2 3B
gpt-oss-20b is the stronger model overall, scoring 32.1 to 28.4 on the Noometry Index.
Last verified . 14 shared benchmarks.
MetaSummary
- They share 14 benchmarks with published results for both. gpt-oss-20b scores higher in 7 categories and Llama 3.2 3B in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where gpt-oss-20b leads 42.4 to 26.4.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.05 / $0.33 for Llama 3.2 3B.
Side by side
| gpt-oss-20b | Llama 3.2 3B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 32.1 | 28.4 |
| Released | 2025-08-05 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 131K | 131K |
| Max output | 16K | 118K |
| Input $ / M tokens | $0.018 | $0.05 |
| Output $ / M tokens | $0.09 | $0.33 |
| Results tracked | 34 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-20b leads
gpt-oss-20b: 37.1 (#185), Llama 3.2 3B: 27.1 (#314)
| Benchmark | gpt-oss-20b | Llama 3.2 3B |
|---|---|---|
| LMArena Coding | 1307 | 1098 |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
| BigCodeBench Instruct | — | 23.4% |
| BigCodeBench Complete | — | 28.3% |
| ALE-Bench | 566.05 | — |
Agentic & Tool Use Llama 3.2 3B leads
gpt-oss-20b: 9.2 (#146), Llama 3.2 3B: 19.9 (#135)
| Benchmark | gpt-oss-20b | Llama 3.2 3B |
|---|---|---|
| Terminal-Bench | 3.4% | — |
| Berkeley Function Calling Leaderboard | — | 21.9% |
| BALROG | — | 10.1% |
Reasoning Llama 3.2 3B leads
gpt-oss-20b: 18.8 (#251), Llama 3.2 3B: 20.1 (#225)
| Benchmark | gpt-oss-20b | Llama 3.2 3B |
|---|---|---|
| LMArena Hard Prompts | 1274 | 1095 |
| Kagi LLM Benchmark | 53.2% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
| Epoch Capabilities Index | 137.81 | — |
Math gpt-oss-20b leads
gpt-oss-20b: 37.6 (#93), Llama 3.2 3B: 29.9 (#213)
| Benchmark | gpt-oss-20b | Llama 3.2 3B |
|---|---|---|
| LMArena Math | 1317 | 1126 |
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| Omni-MATH | 56.5% | — |
Knowledge gpt-oss-20b leads
gpt-oss-20b: 34.6 (#186), Llama 3.2 3B: 29.1 (#232)
| Benchmark | gpt-oss-20b | Llama 3.2 3B |
|---|---|---|
| LMArena Expert | 1258 | 1090 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |
Multilingual gpt-oss-20b leads
gpt-oss-20b: 42.4 (#192), Llama 3.2 3B: 26.4 (#276)
| Benchmark | gpt-oss-20b | Llama 3.2 3B |
|---|---|---|
| LMArena Non-English | 1269 | 1019 |
| LMArena Chinese | 1311 | 1017 |
| LMArena German | 1249 | 1056 |
| LMArena Russian | 1279 | 949 |
| LMArena Japanese | 1245 | — |
| LMArena Korean | 1239 | — |
| LMArena Spanish | 1267 | — |
Instruction Following gpt-oss-20b leads
gpt-oss-20b: 62.0 (#235), Llama 3.2 3B: 56.3 (#270)
| Benchmark | gpt-oss-20b | Llama 3.2 3B |
|---|---|---|
| LMArena Instruction Following | 1236 | 1089 |
| IFEval | 73.2% | — |
Long Context gpt-oss-20b leads
gpt-oss-20b: 37.9 (#203), Llama 3.2 3B: 33.4 (#256)
| Benchmark | gpt-oss-20b | Llama 3.2 3B |
|---|---|---|
| LMArena Longer Query | 1251 | 1101 |
Writing & Preference gpt-oss-20b leads
gpt-oss-20b: 35.8 (#260), Llama 3.2 3B: 24.9 (#302)
| Benchmark | gpt-oss-20b | Llama 3.2 3B |
|---|---|---|
| LMArena Text | 1288 | 1110 |
| LMArena Creative Writing | 1201 | 1094 |
| EQ-Bench Creative Writing | 666 | 595 |
| LMArena Multi-Turn | 1268 | 1105 |
| WildBench | 73.7% | — |
Frequently asked questions
Is gpt-oss-20b better than Llama 3.2 3B?
gpt-oss-20b is the stronger model overall, scoring 32.1 to 28.4 on the Noometry Index.
Which is cheaper, gpt-oss-20b or Llama 3.2 3B?
gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Llama 3.2 3B lists at $0.05 and $0.33.
Is gpt-oss-20b or Llama 3.2 3B better for coding?
gpt-oss-20b scores higher on coding benchmarks: 37.1 versus 27.1 in the Noometry coding category.
Which has the bigger context window?
Both accept 131K tokens.
How many benchmarks do gpt-oss-20b and Llama 3.2 3B share?
14 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Llama 3.2 3B has 18.