In short
How they differ.
Worked out from the figures below. Prices and specs are the makers' own, read Oct 20, 2026.
- DeepSeek-V4-Pro-0813 is cheaper per output token: $3.96 against $10 per million tokens, so GPT-6.1 Sol costs 2.5 times as much.
- GPT-6.1 Sol reads more at once: a context window of 1M tokens against 1M.
- DeepSeek-V4-Pro-0813 can be downloaded and run on your own machine; GPT-6.1 Sol can't.
- On the 1 independent board that rank both, GPT-6.1 Sol places higher on 1 and DeepSeek-V4-Pro-0813 on 0.
Side by side
The specs.
| DeepSeek-V4-Pro-0813 | GPT-6.1 Sol | |
|---|---|---|
| Made by | DeepSeek · China | OpenAI · US |
| Released | Aug 13, 2026 | Sep 29, 2026 |
| Price per 1M tokens | $1.32 in / $3.96 out (peak-hours price) | $2 in / $10 out |
| Context window | 1M | 1M |
| Takes in | text | text, image |
| Download and run it yourself | Yes (MIT) | No |
| Size | 1.6T parameters, 49B active | Not published |
| Maker says | — | “Near-Astra performance for complex work at a lower cost.” |
| Sources | Prices, Oct 20, 2026 | Prices, Oct 20, 2026 · Description |
Independent tests
Where they place.
Position on each board we follow (1 is best). Epoch AI's and Vectara's scores are shown; the other boards are linked for their own figures.
| Board | DeepSeek-V4-Pro-0813 | GPT-6.1 Sol |
|---|---|---|
| Epoch Capabilities IndexEpoch AI, Oct 20, 2026 | Not listed | #3166.32 |
| LMArena coding chatsLMArena, Oct 15, 2026 | Not listed | #10 |
| LMArena WebDevLMArena, Oct 19, 2026 | Not listed | #4 |
| LMArena visionLMArena, Oct 19, 2026 | Not listed | #9 |
| LMArena Agent ArenaLMArena, Oct 19, 2026 | Not listed | #4 |
| Humanity's Last ExamScale AI | #16 | Not listed |
| SWE-Bench ProScale AI, Sep 22, 2026 | Not listed | #7 |
| MultiChallengeScale AI, Feb 3, 2026 | Not listed | #1 |
| Vals IndexVals AI, Oct 19, 2026 | #17 | #7 |
| Terminal-Bench 4.0Vals AI, Oct 19, 2026 | Not listed | #7 |
| Vibe Code BenchVals AI, Oct 18, 2026 | Not listed | #8 |
| ProofBenchVals AI, Oct 12, 2026 | Not listed | #5 |
| GPQA Diamond (run by Vals AI, archived)Vals AI, Sep 1, 2026 | #17 | Not listed |
| Vectara hallucination testVectara, Oct 16, 2026 | #448.6% | Not listed |
More comparisons