Gemini 3.5 Flash vs DeepSeek V4
Google's production / efficiency tier against DeepSeek's open-weight flagship (v4-pro · v4-flash) — published pricing, context windows and specs, side by side.
Gemini 3.5 Flash vs DeepSeek V4: side-by-side
| Gemini 3.5 Flash | DeepSeek V4 | |
|---|---|---|
| Maker | DeepSeek | |
| Released | 19 May 2026 (Google I/O) | 24 April 2026 |
| Tier | Production / efficiency tier | Open-weight flagship (V4-Pro · V4-Flash) |
| Context window | 1,048,576 tokens | 1,000,000 tokens (both tiers) |
| Max output | 65,536 tokens | Not specified |
| API price (per 1M) | $1.50 / $9 per 1M tokens | $1.74 / $3.48 (V4-Pro) · $0.14 / $0.28 (V4-Flash) per 1M |
| Access | Proprietary — Gemini API (Google) | Open weights — MIT licence, on Hugging Face & ModelScope (V4-Pro 1.6T params / 49B active; V4-Flash 284B / 13B active) |
| Best at | Agentic coding, high-volume production work at near-Pro quality | Coding, competitive programming, low-cost agentic work |
| Standout | Beats Gemini 3.1 Pro on coding at 25% lower cost — first public Gemini 3.5 release | Largest open-weights model of 2026 — MIT-licensed 1.6T-param MoE |
Price comparison
| Model | Input (per 1M) | Output (per 1M) |
|---|---|---|
| Gemini 3.5 Flash | $1.50 | $9.00 |
| DeepSeek V4-Flash | $0.14 | $0.28 |
| DeepSeek V4-Pro | $1.74 | $3.48 |
Published benchmark figures
Each figure below is the model's own published score as listed on its AI Model Hub page. Different suites (and suite versions) are not directly comparable.
| Model | Benchmark | Published score |
|---|---|---|
| Gemini 3.5 Flash | Terminal-Bench 2.1 | 76.2% |
| Gemini 3.5 Flash | MCP Atlas | 83.6% |
| Gemini 3.5 Flash | CharXiv Reasoning | 84.2% |
| Gemini 3.5 Flash | GPQA | 82.8% |
| DeepSeek V4 | SWE-bench Verified | 80.6% |
| DeepSeek V4 | Codeforces rating | 3,206 |
What the table says
- Pricing. Gemini 3.5 Flash lists at $1.50 / $9 per 1M tokens; DeepSeek V4 lists at $1.74 / $3.48 (V4-Pro) · $0.14 / $0.28 (V4-Flash) per 1M. On these published rates, Gemini 3.5 Flash's input price is 1.2x lower than DeepSeek V4's, and DeepSeek V4's output price is 2.6x lower than Gemini 3.5 Flash's. Cached input for Gemini 3.5 Flash lists at $0.15 per 1M tokens. DeepSeek V4-Flash lists even lower, at $0.14 / $0.28 per 1M tokens.
- Context. Both are in the same context class: Gemini 3.5 Flash lists 1,048,576 tokens and DeepSeek V4 lists 1,000,000 tokens (both tiers).
- Access. DeepSeek V4 publishes open weights (MIT licence), so it can be self-hosted; Gemini 3.5 Flash is proprietary and available only through hosted APIs and apps.
- Benchmarks. The two model pages report results on different benchmark suites, so the published scores are not directly comparable head-to-head; the tables below list each model's own published figures.
Key takeaway
Everything above is drawn from the two models' published rate cards and spec sheets. Which one fits depends on the workload you price it against — run your own token volumes through the pricing calculator to see the monthly difference.
Full Gemini 3.5 Flash overview Full DeepSeek V4 overview
Frequently asked questions
Which is cheaper, Gemini 3.5 Flash or DeepSeek V4?
On the published rate cards the two are priced identically on the axes compared here; check each model page for tier details.
Which has the bigger context window, Gemini 3.5 Flash or DeepSeek V4?
They are in the same class: Gemini 3.5 Flash lists 1,048,576 tokens and DeepSeek V4 lists 1,000,000 tokens (both tiers).
Which model is newer?
Gemini 3.5 Flash is the more recent release (19 May 2026 (Google I/O)); DeepSeek V4 was released 24 April 2026.
Can I run Gemini 3.5 Flash or DeepSeek V4 locally?
Only DeepSeek V4 — it publishes open weights under the MIT licence. Gemini 3.5 Flash is proprietary and only available via hosted APIs and apps.
Related comparisons
- Gemini 3.5 Flash vs Gemini 3.1 →
- DeepSeek V4 vs GPT-5.5 →
- Kimi K2.6 vs DeepSeek V4 →
- Claude Sonnet 5 vs Gemini 3.5 Flash →