DeepSeek V4 Flash vs Gemini 3.5 Flash: Benchmarks, Pricing & Capabilities Compared
TL;DR — DeepSeek V4 Flash wins for reasoning + cost · Gemini 3.5 Flash wins for general use.
DeepSeek V4 Flash DeepSeek
- Released
- 2026-04-24
- Context window
- 1M tokens
- Input price
- $0.14 / Mtok
- Output price
- $0.28 / Mtok
Gemini 3.5 Flash Google
- Released
- 2026-05-19
- Context window
- 1M tokens
- Input price
- $1.50 / Mtok
- Output price
- $9.00 / Mtok
Benchmark comparison
| Benchmark | DeepSeek V4 Flash | Gemini 3.5 Flash |
|---|---|---|
| AA Intelligence Index | 46.5 ✓ | 43.3 |
| Chatbot Arena Elo | 1433 | 1480 ✓ |
| GPQA Diamond | 89.4% ✓ | 82.8% |
| HLE | 32.1% ✓ | 23.1% |
| IF-Bench | 79.2% ✓ | 47.3% |
| LiveCodeBench Reasoning | 63.0% ✓ | 53.3% |
| SciCode | 44.9% | 48.8% ✓ |
| TAU2-bench | 95.0% ✓ | 58.8% |
| TerminalBench-Hard | 35.6% | 46.2% ✓ |
Pricing comparison
| Metric | DeepSeek V4 Flash | Gemini 3.5 Flash |
|---|---|---|
| Input ($/Mtok) | $0.14 | $1.50 |
| Output ($/Mtok) | $0.28 | $9.00 |
| Cached input ($/Mtok) | — | — |
| Cost per 1M-token roundtrip (1M in + 1M out) | $0.42 | $10.50 |
Context window & modalities
| Attribute | DeepSeek V4 Flash | Gemini 3.5 Flash |
|---|---|---|
| Context window | 1M tokens | 1M tokens |
| Input modalities | text | text, image, audio, video |
| Output modalities | text | text |
| Knowledge cutoff | — | — |
Verdict by use case
Coding
Insufficient data
Basis: SWE-bench
No shared coding benchmark.
Reasoning
→ DeepSeek V4 Flash
Basis: GPQA Diamond
DeepSeek V4 Flash 89.4% vs Gemini 3.5 Flash 82.8% on GPQA Diamond.
Math
Insufficient data
Basis: MATH / AIME
No shared math benchmark.
Long context
Tie
Basis: Context window
DeepSeek V4 Flash 1M tokens vs Gemini 3.5 Flash 1M tokens.
Cost
→ DeepSeek V4 Flash
Basis: Input $/Mtok
DeepSeek V4 Flash $0.14/Mtok vs Gemini 3.5 Flash $1.5/Mtok input.
Changelog & releases
DeepSeek V4 Flash
Released 2026-04-24
Gemini 3.5 Flash
Released 2026-05-19