Model comparison
Chatgpt 4o Latest 20250326 vs DeepSeek V4 Pro
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 43.8 on the Noometry Index.
Last verified . 19 shared benchmarks.
DeepSeekSummary
- They share 19 benchmarks with published results for both. Chatgpt 4o Latest 20250326 scores higher in 0 categories and DeepSeek V4 Pro in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 38.6.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 75% for Chatgpt 4o Latest 20250326 and 53.5% for DeepSeek V4 Pro.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro | |
|---|---|---|
| Provider | OpenAI | DeepSeek |
| Noometry Index | 43.8 | 54.3 |
| Released | — | 2026-04-24 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 393K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1.98 |
| Results tracked | 21 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
Chatgpt 4o Latest 20250326: 41.6 (#122), DeepSeek V4 Pro: 52.4 (#34)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Coding | 1413 | 1470 |
| SWE-bench Verified | — | 77.6% |
| FrontierCode | — | 28.6% |
| LMArena WebDev | — | 1582 |
| SciCode | — | 51% |
| WeirdML | — | 66.2% |
| ALE-Bench | — | 1,403 |
Agentic & Tool Use Not comparable
Chatgpt 4o Latest 20250326: —, DeepSeek V4 Pro: 32.8 (#58)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| APEX-Agents | — | 47.3% |
| Vending-Bench 2 | — | 3,285 |
Reasoning DeepSeek V4 Pro leads
Chatgpt 4o Latest 20250326: 33.8 (#71), DeepSeek V4 Pro: 56.5 (#24)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| Kagi LLM Benchmark | 75% | 53.5% |
| LMArena Hard Prompts | 1424 | 1461 |
| ARC-AGI-2 | — | 61.3% |
| NYT Connections (extended) | — | 91.3% |
| ARC-AGI-1 | — | 90.5% |
| CritPt | — | 18% |
| Chess Puzzles | — | 47% |
| Mystery Game Puzzles | — | 43% |
| DTBench | — | 93.9% |
| LMCA | — | 45.5% |
| Surface Evolver Bench | — | 40% |
| Epoch Capabilities Index | — | 155.31 |
| ForecastBench | — | 56.1 |
Math DeepSeek V4 Pro leads
Chatgpt 4o Latest 20250326: 38.6 (#134), DeepSeek V4 Pro: 64.8 (#30)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Math | 1407 | 1455 |
| FrontierMath (Tiers 1-3) | — | 64.6% |
| FrontierMath Tier 4 | — | 26.8% |
| MathArena Final-Answer Competitions | — | 76.6% |
| OTIS Mock AIME 2024-2025 | — | 98.6% |
| ProofBench | — | 50% |
Knowledge DeepSeek V4 Pro leads
Chatgpt 4o Latest 20250326: 39.2 (#137), DeepSeek V4 Pro: 59.5 (#31)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Expert | 1401 | 1464 |
| GPQA Diamond | — | 91.7% |
| SimpleQA Verified | — | 52.9% |
| Confabulations | 16.6% | — |
| Vectara Hallucination Rate | — | 8.6% |
Multimodal Not comparable
Chatgpt 4o Latest 20250326: 39.6 (#58), DeepSeek V4 Pro: —
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Vision | 1243 | — |
Multilingual DeepSeek V4 Pro leads
Chatgpt 4o Latest 20250326: 52.9 (#76), DeepSeek V4 Pro: 54.4 (#45)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Non-English | 1419 | 1439 |
| LMArena Chinese | 1457 | 1486 |
| LMArena French | 1446 | 1472 |
| LMArena German | 1423 | 1458 |
| LMArena Japanese | 1405 | 1445 |
| LMArena Korean | 1396 | 1447 |
| LMArena Russian | 1429 | 1453 |
| LMArena Spanish | 1434 | 1458 |
Instruction Following DeepSeek V4 Pro leads
Chatgpt 4o Latest 20250326: 74.0 (#107), DeepSeek V4 Pro: 76.1 (#47)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Instruction Following | 1403 | 1448 |
Long Context DeepSeek V4 Pro leads
Chatgpt 4o Latest 20250326: 43.1 (#107), DeepSeek V4 Pro: 45.0 (#51)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Longer Query | 1413 | 1458 |
| CL-bench Life | — | 13.5% |
Writing & Preference DeepSeek V4 Pro leads
Chatgpt 4o Latest 20250326: 62.6 (#72), DeepSeek V4 Pro: 65.5 (#46)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Text | 1429 | 1451 |
| LMArena Creative Writing | 1405 | 1446 |
| EQ-Bench Creative Writing | 1501 | 1553 |
| LMArena Multi-Turn | 1454 | 1467 |
| EQ-Bench 4 | — | 1166 |
Frequently asked questions
Is Chatgpt 4o Latest 20250326 better than DeepSeek V4 Pro?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 43.8 on the Noometry Index.
Is Chatgpt 4o Latest 20250326 or DeepSeek V4 Pro better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 41.6 in the Noometry coding category.
How many benchmarks do Chatgpt 4o Latest 20250326 and DeepSeek V4 Pro share?
19 benchmarks have published results for both models. Chatgpt 4o Latest 20250326 has 21 scored results on Noometry and DeepSeek V4 Pro has 48.