token.app › Compare › GPT-5 Mini vs Gemini 3.7 Flash
GPT-5 Mini vs Gemini 3.7 Flash
Gemini 3.7 Flash leads on 4 of 4 shared benchmarks. GPT-5 Mini is 2.2× cheaper per token on a blended 3:1 basis. Gemini 3.7 Flash accepts the larger context window (1.0M).
Head to head
Blended price uses a 3:1 input:output mix
| GPT-5 Mini | Gemini 3.7 Flash | |
|---|---|---|
| Provider | OpenAI | |
| Input $/1M | $0.250 | $0.750 |
| Output $/1M | $2.00 | $3.75 |
| Blended $/1M | $0.688 | $1.50 |
| Context window | 400K | 1.0M |
| Max output | 128K | 66K |
| Released | Aug 2025 | Aug 2026 |
| Licence | Proprietary | Proprietary |
| GPQA Diamond | 75.0% | 94.8% |
| FrontierMath T1–3 | 46.7% | 71.6% |
| FrontierMath T4 | 12.2% | 36.6% |
| OTIS Mock AIME | 86.7% | 97.2% |
Benchmark scores independently run and published by Epoch AI, 'AI Benchmarking Hub', used under CC BY 4.0. Only benchmarks with a score for BOTH models are compared.
Choose GPT-5 Mini if…
- Costs less per token — $0.688 vs $1.50 blended
Choose Gemini 3.7 Flash if…
- Higher GPQA Diamond (94.8% vs 75.0%)
- Higher FrontierMath T1–3 (71.6% vs 46.7%)
- Higher FrontierMath T4 (36.6% vs 12.2%)
- Higher OTIS Mock AIME (97.2% vs 86.7%)
- Larger context window (1.0M)
- Newer model (released Aug 2026)
FAQ
Which is better, GPT-5 Mini or Gemini 3.7 Flash?
Across the 4 benchmarks both models have independently-run scores for, GPT-5 Mini leads on 0 and Gemini 3.7 Flash leads on 4. Benchmarks are one input — price and context window are on this page too.
Is GPT-5 Mini cheaper than Gemini 3.7 Flash?
GPT-5 Mini is cheaper: $0.688 versus $1.50 per million tokens blended at 3:1 input:output. Input alone: $0.250 vs $0.750. Output alone: $2.00 vs $3.75.
What context windows do GPT-5 Mini and Gemini 3.7 Flash have?
GPT-5 Mini: 400,000 tokens. Gemini 3.7 Flash: 1,048,576 tokens.
Who makes GPT-5 Mini and Gemini 3.7 Flash?
GPT-5 Mini is made by OpenAI. Gemini 3.7 Flash is made by Google.