Model comparison
DeepSeek V4 Pro vs Longcat Flash Chat
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 42.1 on the Noometry Index.
Last verified . 19 shared benchmarks.
DeepSeekSummary
- They share 19 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Longcat Flash Chat in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 19.0.
- The biggest single-benchmark swing is NYT Connections (extended): 91.3% for DeepSeek V4 Pro and 17.7% for Longcat Flash Chat.
Side by side
| DeepSeek V4 Pro | Longcat Flash Chat | |
|---|---|---|
| Provider | DeepSeek | Meituan |
| Noometry Index | 54.3 | 42.1 |
| Released | 2026-04-24 | — |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.66 | — |
| Output $ / M tokens | $1.98 | — |
| Results tracked | 48 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
DeepSeek V4 Pro: 52.4 (#34), Longcat Flash Chat: 43.5 (#87)
| Benchmark | DeepSeek V4 Pro | Longcat Flash Chat |
|---|---|---|
| LMArena Coding | 1470 | 1471 |
| 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
DeepSeek V4 Pro: 32.8 (#58), Longcat Flash Chat: —
| Benchmark | DeepSeek V4 Pro | Longcat Flash Chat |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Longcat Flash Chat: 19.0 (#272)
| Benchmark | DeepSeek V4 Pro | Longcat Flash Chat |
|---|---|---|
| Kagi LLM Benchmark | 53.5% | 43.9% |
| NYT Connections (extended) | 91.3% | 17.7% |
| LMArena Hard Prompts | 1461 | 1440 |
| ARC-AGI-2 | 61.3% | — |
| 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
DeepSeek V4 Pro: 64.8 (#30), Longcat Flash Chat: 39.4 (#107)
| Benchmark | DeepSeek V4 Pro | Longcat Flash Chat |
|---|---|---|
| LMArena Math | 1455 | 1442 |
| 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
DeepSeek V4 Pro: 59.5 (#31), Longcat Flash Chat: 40.6 (#116)
| Benchmark | DeepSeek V4 Pro | Longcat Flash Chat |
|---|---|---|
| LMArena Expert | 1464 | 1454 |
| GPQA Diamond | 91.7% | — |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Longcat Flash Chat: 51.9 (#101)
| Benchmark | DeepSeek V4 Pro | Longcat Flash Chat |
|---|---|---|
| LMArena Non-English | 1439 | 1404 |
| LMArena Chinese | 1486 | 1465 |
| LMArena French | 1472 | 1456 |
| LMArena German | 1458 | 1408 |
| LMArena Japanese | 1445 | 1373 |
| LMArena Korean | 1447 | 1371 |
| LMArena Russian | 1453 | 1395 |
| LMArena Spanish | 1458 | 1445 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Longcat Flash Chat: 74.4 (#96)
| Benchmark | DeepSeek V4 Pro | Longcat Flash Chat |
|---|---|---|
| LMArena Instruction Following | 1448 | 1411 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Longcat Flash Chat: 43.5 (#93)
| Benchmark | DeepSeek V4 Pro | Longcat Flash Chat |
|---|---|---|
| LMArena Longer Query | 1458 | 1425 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Longcat Flash Chat: 61.0 (#91)
| Benchmark | DeepSeek V4 Pro | Longcat Flash Chat |
|---|---|---|
| LMArena Text | 1451 | 1427 |
| LMArena Creative Writing | 1446 | 1388 |
| LMArena Multi-Turn | 1467 | 1418 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Longcat Flash Chat?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 42.1 on the Noometry Index.
Is DeepSeek V4 Pro or Longcat Flash Chat better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 43.5 in the Noometry coding category.
How many benchmarks do DeepSeek V4 Pro and Longcat Flash Chat share?
19 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Longcat Flash Chat has 19.