Model comparison
Gemini 3.1 Flash Lite vs gpt-oss-20b
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 16× less per token, which makes it the better buy when Gemini 3.1 Flash Lite's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
GoogleSummary
- They share 27 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 9 categories and gpt-oss-20b in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemini 3.1 Flash Lite leads 60.9 to 35.5.
- The biggest single-benchmark swing is GPQA Diamond: 81.8% for Gemini 3.1 Flash Lite and 60.8% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.25 / $1.50 for Gemini 3.1 Flash Lite.
- Gemini 3.1 Flash Lite accepts more context: 1.05M tokens versus 131K.
- gpt-oss-20b has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.1 Flash Lite | gpt-oss-20b | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 40.8 | 32.5 |
| Released | 2026-03-03 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 66K | 16K |
| Input $ / M tokens | $0.25 | $0.018 |
| Output $ / M tokens | $1.50 | $0.09 |
| Results tracked | 38 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 3.1 Flash Lite: 37.8 (#188), gpt-oss-20b: 37.6 (#192)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| SciCode | 41.9% | 34.4% |
| WeirdML | 52.2% | 40.9% |
| LMArena Coding | 1400 | 1306 |
| ALE-Bench | 797.73 | 566.05 |
| LMArena WebDev | 1256 | — |
Agentic & Tool Use Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 30.2 (#79), gpt-oss-20b: 9.3 (#154)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
| DeepResearch Bench | 37.3% | — |
Reasoning Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 22.9 (#186), gpt-oss-20b: 19.3 (#261)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| Kagi LLM Benchmark | 67.2% | 53.2% |
| CritPt | 1.1% | 1.4% |
| Chess Puzzles | 25% | 4% |
| LMArena Hard Prompts | 1407 | 1274 |
| DTBench | 76.8% | 68% |
| LMCA | 35% | 14.5% |
| Epoch Capabilities Index | 144.47 | 137.82 |
| NYT Connections (extended) | 8.2% | — |
| EnigmaEval | 3% | — |
| Thematic Generalization | 63.3% | — |
| ForecastBench | 54.4 | — |
Math Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 40.7 (#90), gpt-oss-20b: 39.4 (#103)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 65.3% |
| LMArena Math | 1428 | 1317 |
| FrontierMath (Tiers 1-3) | 27.7% | — |
| Omni-MATH | — | 56.5% |
Knowledge Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 41.9 (#104), gpt-oss-20b: 34.6 (#195)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 81.8% | 60.8% |
| LMArena Expert | 1398 | 1258 |
| Humanity's Last Exam | 8.6% | — |
| MMLU-Pro | — | 74% |
| Vectara Hallucination Rate | 8.2% | — |
| GPQA (HELM) | — | 59.4% |
Multimodal Not comparable
Gemini 3.1 Flash Lite: 39.4 (#60), gpt-oss-20b: —
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| LMArena Vision | 1240 | — |
Multilingual Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 52.3 (#86), gpt-oss-20b: 42.2 (#197)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1411 | 1268 |
| LMArena Chinese | 1461 | 1314 |
| LMArena German | 1429 | 1255 |
| LMArena Japanese | 1413 | 1244 |
| LMArena Korean | 1392 | 1236 |
| LMArena Russian | 1420 | 1278 |
| LMArena Spanish | 1421 | 1267 |
| LMArena French | 1424 | — |
Instruction Following Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 72.7 (#131), gpt-oss-20b: 61.8 (#240)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1377 | 1236 |
| IFEval | — | 73.2% |
Long Context Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 42.5 (#122), gpt-oss-20b: 37.9 (#209)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1394 | 1250 |
Writing & Preference Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 60.9 (#94), gpt-oss-20b: 35.5 (#265)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1416 | 1287 |
| LMArena Creative Writing | 1401 | 1201 |
| LMArena Multi-Turn | 1417 | 1268 |
| EQ-Bench Creative Writing | — | 666 |
| WildBench | — | 73.7% |
Frequently asked questions
Is Gemini 3.1 Flash Lite better than gpt-oss-20b?
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 16× less per token, which makes it the better buy when Gemini 3.1 Flash Lite's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.1 Flash Lite 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; Gemini 3.1 Flash Lite lists at $0.25 and $1.50.
Is Gemini 3.1 Flash Lite or gpt-oss-20b better for coding?
They score almost the same on coding (37.8 vs 37.6); test both on your own repository before choosing.
Which has the bigger context window?
Gemini 3.1 Flash Lite does, with 1.05M tokens against 131K.
How many benchmarks do Gemini 3.1 Flash Lite and gpt-oss-20b share?
27 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and gpt-oss-20b has 34.