DeepSeek V4 Pro vs GPT-6 Sol
DeepSeek V4 Pro 
GPT-6 Sol Verdict
- GPT-6 Sol (max) leads on quality: 67.7 vs 63.5.
- DeepSeek V4 Pro (max) is stronger in agents & tools, coding, instruction following, maths.
- GPT-6 Sol (max) is stronger in composite, knowledge, long context, reasoning, multimodal.
- DeepSeek V4 Pro (max) is 7.4× cheaper ($0.544 vs $4.00 per 1M blended).
- GPT-6 Sol (max) streams 1.5× faster (126 vs 86 tokens per second).
| Metric | DeepSeek V4 Pro (max) | GPT-6 Sol (max) |
|---|---|---|
| BenchLeader Index | 63.5 | 67.7 |
| Agents & tools score | 63.6 | – |
| Coding score | 59.2 | – |
| Composite score | 72.4 | 86.2 |
| Instruction following score | 74.3 | – |
| Knowledge score | 60.2 | 75.4 |
| Long context score | 66.0 | 67.7 |
| Maths score | 57.1 | – |
| Reasoning score | 66.8 | 95.0 |
| Multimodal score | – | 67.1 |
| Blended price $/M | $0.544 | $4.00 |
| Output speed | 86 tok/s | 126 tok/s |
| Time to first answer | 52.5 s | 107.2 s |
| Context window | 1M | 1.1M |
| GPQA Diamond | 91.7% | – |
| FrontierMath Tiers 1–3 | 64.6% | – |
| FrontierMath Tier 4 | 26.8% | – |
| OTIS Mock AIME | 98.6% | – |
| SWE-bench Verified (Epoch) | 77.6% | – |
| SimpleQA Verified | 52.9% | – |
| SciCode | 51.0% | – |
| WeirdML | 66.2% | – |
| ProofBench | 16.0% | – |
| AA Intelligence Index | 36 | 47.5 |
| IFBench | 76.5% | – |
| AA-LCR | 80.3% | 83.7% |
| MMMU-Pro | – | 83.3% |
| AA-Omniscience | 0.8 | 27.1 |
| Terminal-Bench Hard | 46.2% | – |
| GPQA Diamond (AA) | 92.8% | – |
| Humanity's Last Exam (AA) | 41.0% | 47.9% |
| SciCode (AA) | 51.0% | 57.6% |
| τ²-Bench Telecom (AA) | 96.2% | – |
| LiveCodeBench | 87.5% | – |
| MMLU-Pro | 87.3% | – |
| IOI | 51.6% | – |
| LegalBench | 82.4% | – |
| CorpFin | 65.4% | – |
| TaxEval | 73.1% | – |
| Terminal-Bench 2.1 (Vals) | 54.7% | – |
| SWE-bench (Vals) | 96.4% | – |
| GPQA Diamond (Vals) | 92.4% | – |
| Vals Index | 52.4 | – |
| AIME 2026 | 96.7% | – |
| HMMT February 2026 | 93.9% | – |
| MathArena Apex | 28.1% | – |
| ARC-AGI-1 | 90.0% | – |
| ARC-AGI-2 | 61.3% | – |
| CritPt | 18.0% | 30.9% |
| GDPval (AA) | 47.1% | 49.4% |
| τ²-Bench Banking (AA) | 39.6% | – |
| ITBench SRE (AA) | 38.3% | – |
| Analyst Agent (AA) | 18.8% | – |
| APEX-Agents (AA) | 24.3% | – |
| CaseLaw v2 | 59.4% | – |
| Code Migration | 41.5% | – |
| Excel Modeling Benchmark | 52.8% | – |
| Finance Agent v2 | 50.4% | – |
| Harvey's Legal Agent Benchmark | 7.5% | – |
| Legal Research Bench | 40.9% | – |
| MedCode | 42.5% | – |
| MedScribe | 80.2% | – |
| ProgramBench | 0.0% | – |
| Public Benefits Bench | 62.9% | – |
| SkillsBench | 53.8% | – |
| Tax Agent Bench | 58.7% | – |
| Terminal-Bench 4.0 (Vals) | 1.0% | – |
| Terminal-Bench Science | 0.0% | – |
| Vibe Code Bench 1-100 | 17.5% | – |
| Vibe Code Bench v1.1 | 82.3% | – |
| Chess Puzzles | 47.0% | – |
| Mystery Game Puzzles | 43.0% | – |
| LMCA | 41.2% | – |
| DTBench | 90.7% | – |
| ALE-Bench | 1403.2 | – |
| Terminal-Bench 4.0 (AA) | 14.7% | 43.9% |
| Terminal-Bench 2.1 (AA) | 78.7% | – |
| AutomationBench | 56.7% | 61.6% |
| GDP.pdf | 11.4% | 24.8% |
| MLCR | 17.8% | 16.1% |
| EnterpriseOps-Gym | 49.6% | – |
| AA-Omniscience: accuracy | 49.1% | 54.5% |
| AA-Omniscience: non-hallucination | 5.9% | 39.9% |
| AA-Briefcase | 1261 | 1483 |
| AA Openness Index | 44.4 | – |
Data as of 2026-09-23. Best configuration of each model; every score links to its source on the model pages.
DeepSeek V4 Pro vs GPT-6 Sol: questions
- Is DeepSeek V4 Pro better than GPT-6 Sol?
- GPT-6 Sol (max) leads on quality: 67.7 vs 63.5. The BenchLeader Index combines every independent quality benchmark; GPT-6 Sol (max) is ahead overall as of 2026-09-23, but check the category scores for your use.
- Which is cheaper, DeepSeek V4 Pro or GPT-6 Sol?
- DeepSeek V4 Pro is cheaper: $0.544 against $4.00 per million tokens, blended at three input tokens per output token.
- Which is faster, DeepSeek V4 Pro or GPT-6 Sol?
- GPT-6 Sol streams faster: 126 against 86 output tokens per second.
- Which has the larger context window?
- GPT-6 Sol accepts more context: 1.1M against 1M tokens.