Model comparison
AI Co-Mathematician vs DeepSeek V4 Pro
DeepSeek V4 Pro has enough public results to be ranked (#31); AI Co-Mathematician does not yet, so treat this comparison as directional.
Last verified . 1 shared benchmarks.
Google
DeepSeekSummary
- They share 1 benchmark with published results for both. AI Co-Mathematician scores higher in 1 category and DeepSeek V4 Pro in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in math, where AI Co-Mathematician leads 72.3 to 64.8.
- The biggest single-benchmark swing is FrontierMath Tier 4: 75.6% for AI Co-Mathematician and 26.8% for DeepSeek V4 Pro.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| AI Co-Mathematician | DeepSeek V4 Pro | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 58.1 | 54.3 |
| Released | 2026-05-08 | 2026-04-24 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 393K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1.98 |
| Results tracked | 2 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
AI Co-Mathematician: —, DeepSeek V4 Pro: 52.4 (#34)
| Benchmark | AI Co-Mathematician | DeepSeek V4 Pro |
|---|---|---|
| SWE-bench Verified | — | 77.6% |
| FrontierCode | — | 28.6% |
| LMArena WebDev | — | 1582 |
| SciCode | — | 51% |
| WeirdML | — | 66.2% |
| LMArena Coding | — | 1470 |
| ALE-Bench | — | 1,403 |
Agentic & Tool Use Not comparable
AI Co-Mathematician: —, DeepSeek V4 Pro: 32.8 (#58)
| Benchmark | AI Co-Mathematician | DeepSeek V4 Pro |
|---|---|---|
| APEX-Agents | — | 47.3% |
| Vending-Bench 2 | — | 3,285 |
Reasoning Not comparable
AI Co-Mathematician: —, DeepSeek V4 Pro: 56.5 (#24)
| Benchmark | AI Co-Mathematician | DeepSeek V4 Pro |
|---|---|---|
| ARC-AGI-2 | — | 61.3% |
| Kagi LLM Benchmark | — | 53.5% |
| NYT Connections (extended) | — | 91.3% |
| ARC-AGI-1 | — | 90.5% |
| CritPt | — | 18% |
| Chess Puzzles | — | 47% |
| LMArena Hard Prompts | — | 1461 |
| Mystery Game Puzzles | — | 43% |
| DTBench | — | 93.9% |
| LMCA | — | 45.5% |
| Surface Evolver Bench | — | 40% |
| Epoch Capabilities Index | — | 155.31 |
| ForecastBench | — | 56.1 |
Math AI Co-Mathematician leads
AI Co-Mathematician: 72.3, DeepSeek V4 Pro: 64.8 (#30)
| Benchmark | AI Co-Mathematician | DeepSeek V4 Pro |
|---|---|---|
| FrontierMath Tier 4 | 75.6% | 26.8% |
| FrontierMath (Tiers 1-3) | — | 64.6% |
| MathArena Final-Answer Competitions | — | 76.6% |
| OTIS Mock AIME 2024-2025 | — | 98.6% |
| ProofBench | — | 50% |
| LMArena Math | — | 1455 |
| FrontierMath Tier 4 (v1) | 47.9% | — |
Knowledge Not comparable
AI Co-Mathematician: —, DeepSeek V4 Pro: 59.5 (#31)
| Benchmark | AI Co-Mathematician | DeepSeek V4 Pro |
|---|---|---|
| GPQA Diamond | — | 91.7% |
| SimpleQA Verified | — | 52.9% |
| Vectara Hallucination Rate | — | 8.6% |
| LMArena Expert | — | 1464 |
Multilingual Not comparable
AI Co-Mathematician: —, DeepSeek V4 Pro: 54.4 (#45)
| Benchmark | AI Co-Mathematician | DeepSeek V4 Pro |
|---|---|---|
| LMArena Non-English | — | 1439 |
| LMArena Chinese | — | 1486 |
| LMArena French | — | 1472 |
| LMArena German | — | 1458 |
| LMArena Japanese | — | 1445 |
| LMArena Korean | — | 1447 |
| LMArena Russian | — | 1453 |
| LMArena Spanish | — | 1458 |
Instruction Following Not comparable
AI Co-Mathematician: —, DeepSeek V4 Pro: 76.1 (#47)
| Benchmark | AI Co-Mathematician | DeepSeek V4 Pro |
|---|---|---|
| LMArena Instruction Following | — | 1448 |
Long Context Not comparable
AI Co-Mathematician: —, DeepSeek V4 Pro: 45.0 (#51)
| Benchmark | AI Co-Mathematician | DeepSeek V4 Pro |
|---|---|---|
| CL-bench Life | — | 13.5% |
| LMArena Longer Query | — | 1458 |
Writing & Preference Not comparable
AI Co-Mathematician: —, DeepSeek V4 Pro: 65.5 (#46)
| Benchmark | AI Co-Mathematician | DeepSeek V4 Pro |
|---|---|---|
| LMArena Text | — | 1451 |
| LMArena Creative Writing | — | 1446 |
| EQ-Bench Creative Writing | — | 1553 |
| EQ-Bench 4 | — | 1166 |
| LMArena Multi-Turn | — | 1467 |
Frequently asked questions
Is AI Co-Mathematician better than DeepSeek V4 Pro?
DeepSeek V4 Pro has enough public results to be ranked (#31); AI Co-Mathematician does not yet, so treat this comparison as directional.
How many benchmarks do AI Co-Mathematician and DeepSeek V4 Pro share?
1 benchmark has published results for both models. AI Co-Mathematician has 2 scored results on Noometry and DeepSeek V4 Pro has 48.