DeepSeek V3.1 vs DeepSeek V4 Pro 0813
| Benchmark | DeepSeek V3.1 | DeepSeek V4 Pro 0813 |
|---|---|---|
| Humanity's Last Exam | 6.7 | 10.6 |
| Terminal-Bench 4.0 | — | 1.5 |
| AutomationBench-AA | — | 34.9 |
Specs
| Spec | DeepSeek V3.1 | DeepSeek V4 Pro 0813 |
|---|---|---|
| Input, per 1M | $0.25 | $1.32 |
| Output, per 1M | $0.95 | $3.96 |
| Cost per task? | — | $0.36 |
| Speed, tok/s | 98 | 104 |
| Context | 128K | 1M |
| Weights | Open | Open |
| Released | Aug 2025 | Aug 2026 |
| Memory (4-bit) | 411 GB | 990 GB |
| Licence | MIT | MIT |
| Commercial use | Allowed | Allowed |
| See images | — | — |
| Read PDFs | Partial | — |
| Hear audio | — | — |
| Watch video | — | — |
| Call tools | ✓ | ✓ |
| Force a tool call | Partial | ✓ |
| Return strict JSON | ✓ | ✓ |
| Search the web | ? | ? |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
DeepSeek V3.1 → DeepSeek V4 Pro 0813
Memory (4-bit): 411 GB → 990 GB
Pass back reasoning_content · Chat Completions›
With thinking and tools, return all prior reasoning_content even on turns without tool calls or the API returns HTTP 400.
Disable thinking mode first · Chat Completions›
Required and named tool choices return HTTP 400 in thinking mode; disable thinking first.
Parameters have no effect›
In thinking mode, temperature, presence_penalty and frequency_penalty have no effect.
top_p value is ignored›
In thinking mode, top_p is fixed at 0.95 with tools and 1.0 otherwise; in non-thinking mode it takes effect.
Stateless API · Responses›
Responses does not support previous_response_id or conversation; preserve conversation history in your client.
DeepSeek V4 Pro 0813 → DeepSeek V3.1