Model comparison
Gemini 3.1 Flash Lite vs gpt-oss-120b
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.4 to 36.0 on the Noometry Index. gpt-oss-120b costs 8.0× 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 . 29 shared benchmarks.
GoogleSummary
- They share 29 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 7 categories and gpt-oss-120b in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Gemini 3.1 Flash Lite leads 29.9 to 12.1.
- The biggest single-benchmark swing is LMCA: 35% for Gemini 3.1 Flash Lite and 22.1% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 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-120b has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.1 Flash Lite | gpt-oss-120b | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 40.4 | 36.0 |
| Released | 2026-03-03 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 66K | 41K |
| Input $ / M tokens | $0.25 | $0.037 |
| Output $ / M tokens | $1.50 | $0.17 |
| Results tracked | 38 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 37.5 (#179), gpt-oss-120b: 33.2 (#251)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| SciCode | 41.9% | 36% |
| WeirdML | 52.2% | 48.2% |
| LMArena Coding | 1400 | 1380 |
| ALE-Bench | 797.73 | 575.62 |
| SWE-bench Verified (bash only) | — | 26% |
| Aider Polyglot | — | 41.8% |
| LMArena WebDev | 1256 | — |
| AlgoTune | — | 1.41 |
Agentic & Tool Use Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 29.9 (#72), gpt-oss-120b: 12.1 (#145)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| Terminal-Bench | — | 18.7% |
| APEX-Agents | — | 4.4% |
| DeepResearch Bench | 37.3% | — |
| METR Time Horizons | — | 56.6% |
| Vending-Bench 2 | — | -21.53 |
Reasoning Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 22.3 (#181), gpt-oss-120b: 19.5 (#236)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| Kagi LLM Benchmark | 67.2% | 58.6% |
| CritPt | 1.1% | 1.1% |
| Chess Puzzles | 25% | 20% |
| LMArena Hard Prompts | 1407 | 1364 |
| DTBench | 76.8% | 76.3% |
| LMCA | 35% | 22.1% |
| Epoch Capabilities Index | 144.45 | 139.94 |
| SimpleBench | — | 22.1% |
| NYT Connections (extended) | 8.2% | — |
| EnigmaEval | 3% | — |
| Thematic Generalization | 63.3% | — |
| Mystery Game Puzzles | — | 2% |
| Surface Evolver Bench | — | 25% |
| ForecastBench | 54.4 | — |
Math gpt-oss-120b leads
Gemini 3.1 Flash Lite: 39.5 (#82), gpt-oss-120b: 50.6 (#43)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 88.9% |
| LMArena Math | 1427 | 1388 |
| FrontierMath (Tiers 1-3) | 27.7% | — |
| Omni-MATH | — | 68.8% |
Knowledge Too close to call
Gemini 3.1 Flash Lite: 41.7 (#95), gpt-oss-120b: 42.3 (#89)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 81.8% | 75.8% |
| Vectara Hallucination Rate | 8.2% | 14.2% |
| LMArena Expert | 1397 | 1356 |
| Humanity's Last Exam | 8.6% | — |
| MMLU-Pro | — | 79.5% |
| Confabulations | — | 15.7% |
| GPQA (HELM) | — | 68.4% |
Multimodal Not comparable
Gemini 3.1 Flash Lite: 38.7 (#57), gpt-oss-120b: —
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| LMArena Vision | 1240 | — |
Multilingual Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 52.5 (#82), gpt-oss-120b: 48.2 (#142)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1410 | 1350 |
| LMArena Chinese | 1461 | 1384 |
| LMArena French | 1423 | 1370 |
| LMArena German | 1429 | 1355 |
| LMArena Japanese | 1411 | 1332 |
| LMArena Korean | 1393 | 1283 |
| LMArena Russian | 1420 | 1343 |
| LMArena Spanish | 1420 | 1388 |
Instruction Following Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 72.9 (#126), gpt-oss-120b: 69.5 (#168)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| LMArena Instruction Following | 1376 | 1318 |
| IFEval | — | 83.6% |
Long Context Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 42.5 (#117), gpt-oss-120b: 31.4 (#273)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| LMArena Longer Query | 1394 | 1319 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 61.0 (#90), gpt-oss-120b: 46.7 (#213)
| Benchmark | Gemini 3.1 Flash Lite | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1416 | 1366 |
| LMArena Creative Writing | 1401 | 1274 |
| LMArena Multi-Turn | 1417 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
| EQ-Bench Creative Writing | — | 961 |
| WildBench | — | 84.5% |
Frequently asked questions
Is Gemini 3.1 Flash Lite better than gpt-oss-120b?
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.4 to 36.0 on the Noometry Index. gpt-oss-120b costs 8.0× 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-120b?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 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-120b better for coding?
Gemini 3.1 Flash Lite scores higher on coding benchmarks: 37.5 versus 33.2 in the Noometry coding category.
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-120b share?
29 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and gpt-oss-120b has 48.