Model comparison
Chatgpt 4o Latest 20250326 vs DeepSeek V4 Flash
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 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 Flash in 8 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Flash leads 60.3 to 38.6.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 75% for Chatgpt 4o Latest 20250326 and 52.2% for DeepSeek V4 Flash.
- DeepSeek V4 Flash has downloadable open weights; the other is API-only.
Side by side
| Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash | |
|---|---|---|
| Provider | OpenAI | DeepSeek |
| Noometry Index | 43.8 | 53.6 |
| Released | — | 2026-04-24 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 393K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 21 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Flash leads
Chatgpt 4o Latest 20250326: 41.6 (#122), DeepSeek V4 Flash: 47.9 (#59)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Coding | 1413 | 1457 |
| FrontierCode | — | 18.8% |
| LMArena WebDev | — | 1582 |
| SciCode | — | 49.9% |
| WeirdML | — | 63% |
| ALE-Bench | — | 1,306 |
Reasoning DeepSeek V4 Flash leads
Chatgpt 4o Latest 20250326: 33.8 (#71), DeepSeek V4 Flash: 53.7 (#30)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash |
|---|---|---|
| Kagi LLM Benchmark | 75% | 52.2% |
| LMArena Hard Prompts | 1424 | 1444 |
| ARC-AGI-2 | — | 61.4% |
| SimpleBench | — | 61.1% |
| NYT Connections (extended) | — | 89.6% |
| ARC-AGI-1 | — | 89% |
| CritPt | — | 16.6% |
| Chess Puzzles | — | 33% |
| Mystery Game Puzzles | — | 34% |
| DTBench | — | 90.9% |
| LMCA | — | 41.7% |
| Epoch Capabilities Index | — | 154.49 |
Math DeepSeek V4 Flash leads
Chatgpt 4o Latest 20250326: 38.6 (#134), DeepSeek V4 Flash: 60.3 (#37)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Math | 1407 | 1427 |
| FrontierMath (Tiers 1-3) | — | 57.5% |
| FrontierMath Tier 4 | — | 24.4% |
| MathArena Final-Answer Competitions | — | 76.5% |
| OTIS Mock AIME 2024-2025 | — | 94.4% |
| ProofBench | — | 56% |
Knowledge DeepSeek V4 Flash leads
Chatgpt 4o Latest 20250326: 39.2 (#137), DeepSeek V4 Flash: 55.4 (#48)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Expert | 1401 | 1441 |
| GPQA Diamond | — | 91% |
| SimpleQA Verified | — | 33.6% |
| Confabulations | 16.6% | — |
Multimodal Not comparable
Chatgpt 4o Latest 20250326: 39.6 (#58), DeepSeek V4 Flash: —
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Vision | 1243 | — |
Multilingual Too close to call
Chatgpt 4o Latest 20250326: 52.9 (#76), DeepSeek V4 Flash: 53.0 (#72)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Non-English | 1419 | 1420 |
| LMArena Chinese | 1457 | 1468 |
| LMArena French | 1446 | 1439 |
| LMArena German | 1423 | 1418 |
| LMArena Japanese | 1405 | 1406 |
| LMArena Korean | 1396 | 1384 |
| LMArena Russian | 1429 | 1428 |
| LMArena Spanish | 1434 | 1436 |
Instruction Following Too close to call
Chatgpt 4o Latest 20250326: 74.0 (#107), DeepSeek V4 Flash: 74.9 (#81)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Instruction Following | 1403 | 1421 |
Long Context Too close to call
Chatgpt 4o Latest 20250326: 43.1 (#107), DeepSeek V4 Flash: 43.8 (#85)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Longer Query | 1413 | 1434 |
Writing & Preference DeepSeek V4 Flash leads
Chatgpt 4o Latest 20250326: 62.6 (#72), DeepSeek V4 Flash: 63.8 (#61)
| Benchmark | Chatgpt 4o Latest 20250326 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Text | 1429 | 1432 |
| LMArena Creative Writing | 1405 | 1403 |
| EQ-Bench Creative Writing | 1501 | 1559 |
| LMArena Multi-Turn | 1454 | 1449 |
Frequently asked questions
Is Chatgpt 4o Latest 20250326 better than DeepSeek V4 Flash?
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 43.8 on the Noometry Index.
Is Chatgpt 4o Latest 20250326 or DeepSeek V4 Flash better for coding?
DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 41.6 in the Noometry coding category.
How many benchmarks do Chatgpt 4o Latest 20250326 and DeepSeek V4 Flash share?
19 benchmarks have published results for both models. Chatgpt 4o Latest 20250326 has 21 scored results on Noometry and DeepSeek V4 Flash has 41.