DeepSeek-V4-Flash-Vision-ExpvsGemini 3.7 Flash
Across 4 shared benchmarks, Gemini 3.7 Flash leads overall: DeepSeek-V4-Flash-Vision-Exp wins 1, Gemini 3.7 Flash wins 3, with 0 ties and an average score difference of -2.90.
DeepSeek-V4-Flash-Vision-Exp
DeepSeek-AI · 2026-08-21 · Multimodal model
Gemini 3.7 Flash
Google Deep Mind · 2026-08-13 · Multimodal model
DeepSeek-V4-Flash-Vision-Exp1 win(25%)(75%)3 winsGemini 3.7 Flash
Benchmark scores
Grouped by capability, sorted by largest gap within each. 4 shared benchmarks.
AI Agent - Tool Usage
Gemini 3.7 Flash 2/2| Benchmark | DeepSeek-V4-Flash-Vision-Exp | Gemini 3.7 Flash | Diff |
|---|---|---|---|
| AutomationBench | 25.707 / 8Max (With Tools) | 30.404 / 8Thinking (With Tools) | -4.70 |
| Terminal-Bench 2.1 | 83.9011 / 44Max (With Tools) | 85.809 / 44Thinking (With Tools) | -1.90 |
Agent Level Benchmark
DeepSeek-V4-Flash-Vision-Exp 1/1| Benchmark | DeepSeek-V4-Flash-Vision-Exp | Gemini 3.7 Flash | Diff |
|---|---|---|---|
| Agents' Last Exam | 27.306 / 11Max (With Tools) | 26.308 / 11Thinking Medium (With Tools) | +1 |
Coding and Software Engineer
Gemini 3.7 Flash 1/1| Benchmark | DeepSeek-V4-Flash-Vision-Exp | Gemini 3.7 Flash | Diff |
|---|---|---|---|
| DeepSWE | 59.3012 / 27Max (With Tools) | 65.3010 / 27Thinking High (With Tools) | -6 |
Specs
| Field | DeepSeek-V4-Flash-Vision-Exp | Gemini 3.7 Flash |
|---|---|---|
| Publisher | DeepSeek-AI | Google Deep Mind |
| Release date | 2026-08-21 | 2026-08-13 |
| Model type | Multimodal model | Multimodal model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1M | 1M |
| Max output | 384K | 64K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | DeepSeek-V4-Flash-Vision-Exp | Gemini 3.7 Flash |
|---|---|---|
| Text input | $0.22 / 1M tokens | $0.75 / 1M tokens |
| Text output | $0.66 / 1M tokens | $3.75 / 1M tokens |
| Cache read | $0.007 / 1M tokens | $0.075 / 1M tokens |
Summary
- DeepSeek-V4-Flash-Vision-Expleads in:Agent Level Benchmark (1/1)
- Gemini 3.7 Flashleads in:AI Agent - Tool Usage (2/2), Coding and Software Engineer (1/1)
On average across the 4 shared benchmarks, Gemini 3.7 Flash scores 2.90 higher.
Largest single-benchmark gap: DeepSWE — DeepSeek-V4-Flash-Vision-Exp 59.30 vs Gemini 3.7 Flash 65.30 (-6).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.