Model comparison
gpt-oss-20b vs Gemini 2.5 Flash-Lite
- Input
- $0.029 / 1M tokens
- Output
- $0.14 / 1M tokens
- Context
- 131K
- Released
- 2025-08-05
- Input
- $0.10 / 1M tokens (text / image / video)
- Output
- $0.40 / 1M tokens
- Context
- 1M
- Released
- 2025-06-17
Gpt-oss-20b scores higher on 4 of the 4 benchmarks both models report; gpt-oss-20b is cheaper on input tokens ($0 vs $0.1 per 1M, 3.4×); Gemini 2.5 Flash-Lite has the larger context window (1M vs 131K); gpt-oss-20b ships open weights while Gemini 2.5 Flash-Lite is API-only.
Specs and pricing side by side
| Spec | gpt-oss-20b | Gemini 2.5 Flash-Lite |
|---|---|---|
| Maker | OpenAI | |
| Released | 2025-08-05 | 2025-06-17 |
| Context window | 131K | 1M |
| Max output | 131K | 64K |
| Input price | $0.029 / 1M tokens | $0.10 / 1M tokens (text / image / video) |
| Output price | $0.14 / 1M tokens | $0.40 / 1M tokens |
| Parameters | 20.9B total / 3.6B active (MoE) | Undisclosed |
| License | Apache 2.0 | Proprietary (Google), API-only |
| Weights | Open weights | API only |
| Modalities | Text | Text, Vision, Audio, Video, PDF |
| Knowledge cutoff | June 2024 | January 2025 |
Benchmarks both models report
| Benchmark | gpt-oss-20b | Gemini 2.5 Flash-Lite |
|---|---|---|
| GPQA Diamond | 71.5% | 64.6% |
| AIME 2025 | 98.7% | 49.8% |
| SWE-bench Verified | 60.7% | 31.6% |
| Humanity's Last Exam | 17.3% | 5.1% |
Scores are as published by each maker or its cited source; harness, reasoning effort and date can differ between rows, so treat small gaps as ties. Sources are listed on each model page.
Which one should you pick?
Pick gpt-oss-20b if…
- You care about GPQA Diamond, AIME 2025, SWE-bench Verified (it scores higher on 4 shared benchmarks).
- Cost matters: input tokens cost 3.4× less ($0 vs $0.1 per 1M).
- You want to self-host or fine-tune (open weights, Apache 2.0).
- Your use case is: on-device and local reasoning assistants that keep data off the cloud.
Pick Gemini 2.5 Flash-Lite if…
- You need the longer context (1M vs 131K).
- You need Vision / Audio / Video / PDF input, which gpt-oss-20b lacks.
- Your use case is: high-volume classification and intelligent request routing.
FAQ
Which is better, gpt-oss-20b or Gemini 2.5 Flash-Lite?
Gpt-oss-20b scores higher on 4 of the 4 benchmarks both models report; gpt-oss-20b is cheaper on input tokens ($0 vs $0.1 per 1M, 3.4×); Gemini 2.5 Flash-Lite has the larger context window (1M vs 131K); gpt-oss-20b ships open weights while Gemini 2.5 Flash-Lite is API-only. The better pick depends on whether you value benchmark scores, price per token, context length or openness for your workload.
How much cheaper is gpt-oss-20b?
gpt-oss-20b: $0.029 / 1M tokens input, $0.14 / 1M tokens output. Gemini 2.5 Flash-Lite: $0.10 / 1M tokens (text / image / video) input, $0.40 / 1M tokens output (list API prices as published by each maker).
What are the context windows of gpt-oss-20b and Gemini 2.5 Flash-Lite?
gpt-oss-20b offers 131K with up to 131K output; Gemini 2.5 Flash-Lite offers 1M with up to 64K output.
Is gpt-oss-20b newer than Gemini 2.5 Flash-Lite?
gpt-oss-20b was released 2025-08-05; Gemini 2.5 Flash-Lite was released 2025-06-17.
Go deeper
- gpt-oss-20b: full specs, pricing and benchmarks
- Gemini 2.5 Flash-Lite: full specs, pricing and benchmarks
- gemini 2 5 flash lite vs gemini 2 0 flash lite
- gemini 3 1 flash lite vs gemini 2 5 flash lite
- gpt oss 20b vs kimi k2
- deepseek v3 1 terminus vs gemini 2 5 flash lite
- gpt oss 20b vs gemini 2 5 pro
- gpt oss 20b vs claude opus 4
- All model comparisons