DeepSeek V3.1 vs DeepSeek V4 Flash Vision Exp
DeepSeek V3.1 (DeepSeek) and DeepSeek V4 Flash Vision Exp (DeepSeek) are listed at different prices. DeepSeek V4 Flash Vision Exp costs less: $0.216 per 1M input tokens and $0.647 per 1M output tokens, against $0.25 and $0.95 for DeepSeek V3.1. DeepSeek V4 Flash Vision Exp has the larger context window, 1,048,576 tokens against 163,840 for DeepSeek V3.1. Only DeepSeek V4 Flash Vision Exp supports a reasoning effort setting. DeepSeek V3.1 accepts text and produces text; DeepSeek V4 Flash Vision Exp accepts text and images and produces text.
DeepSeek: DeepSeek V3.1 accepts and produces text, while DeepSeek: DeepSeek V4 Flash Vision Exp accepts text and images and produces text. Both support tools, structured outputs, response formats and reasoning; V4 Flash Vision Exp has a 1,048,576-token context, compared with V3.1’s 163,840 tokens.
Compared from catalog data checked
Side by side
0.00 the better figure in a row| Specification | DeepSeek V3.1 | DeepSeek V4 Flash Vision Exp |
|---|---|---|
| Maker | DeepSeek | DeepSeek |
| Added to catalog | 21 Aug 2025 | 21 Aug 2026 |
| Input, per 1M tokens | $0.25 | $0.216 |
| Output, per 1M tokens | $0.95 | $0.647 |
| Cached input read, per 1M | $0.13 | $0.0069 |
| Context windowTokens | 163,840 | 1,048,576 |
| Max outputTokens | 147,456 | 262,144 |
| Accepts | text | image, text |
| Produces | text | text |
| Tool calling | Yes | Yes |
| Structured outputs | Yes | Yes |
| Reasoning effort control | No | Yes |
List prices through OpenRouter, in US dollars. Each model's page has its variants, price history and sample outputs.
What the same task costs
List price for one request, from the token counts shown. The same four tasks are priced on every model page.
| Task | DeepSeek V3.1 | DeepSeek V4 Flash Vision Exp |
|---|---|---|
| Answer a short question500 in / 300 out | $0.0004 | $0.0003 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0070 | $0.0059 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0094 | $0.0078 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.039 | $0.033 |
Which to choose
Choose DeepSeek V3.1 if
No row of the table favours DeepSeek V3.1. The difference, if there is one, is in the answers it writes: try it on your own task.
DeepSeek V3.1 detailsChoose DeepSeek V4 Flash Vision Exp if
- Input tokens cost less: $0.216 against $0.25 per 1M
- Output tokens cost less: $0.647 against $0.95 per 1M
- Cached input costs less: $0.0069 against $0.13 per 1M
- You need the larger context window: 1,048,576 tokens against 163,840
- You need longer answers in one response: 262,144 tokens against 147,456
- It has to read images, which the other model does not
- You want to set how much it reasons before it answers
Neither maker's quality claims are repeated here. The reliable test is your own task: ask both the same question and read the two answers.