# DeepSeek V4.1 Flash vs Llama 4 Maverick

> DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 30.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-1-flash-vs-llama-4-maverick
- Last updated: 2026-10-12
- Shared benchmarks: 28

## Summary

- They share 28 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 10 categories and Llama 4 Maverick in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 26.0.
- The biggest single-benchmark swing is NYT Connections (extended): 89.6% for DeepSeek V4.1 Flash and 8% for Llama 4 Maverick.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.19 / $0.65 for Llama 4 Maverick.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 128K.

## Snapshot

| | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 52.8 | 30.9 |
| Rank | 38 | 282 |
| Context | 1M | 128K |
| Input $/M | $0.15 | $0.19 |
| Output $/M | $0.60 | $0.65 |
| Weights | Open | Open |

## Coding

- DeepSeek V4.1 Flash: 52.9 (#32)
- Llama 4 Maverick: 26.6 (#324)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| SciCode | 51.9% | 33.1% |
| LMArena Coding | 1506 | 1302 |
| ALE-Bench | 1,092 | 172.97 |
| SWE-bench Verified (bash only) | — | 21% |
| Aider Polyglot | — | 15.6% |
| LMArena WebDev | 1619 | — |
| WeirdML | — | 24.5% |
| BigCodeBench Instruct | — | 49.7% |
| BigCodeBench Complete | — | 61.4% |

## Agentic & Tool Use

- DeepSeek V4.1 Flash: 31.2 (#69)
- Llama 4 Maverick: 28.2 (#91)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| APEX-Agents | 39.5% | — |
| Berkeley Function Calling Leaderboard | — | 37.3% |
| GDP.pdf | 19.8% | — |

## Reasoning

- DeepSeek V4.1 Flash: 50.2 (#36)
- Llama 4 Maverick: 10.1 (#342)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| NYT Connections (extended) | 89.6% | 8% |
| CritPt | 14.3% | 0% |
| LMArena Hard Prompts | 1483 | 1281 |
| DTBench | 89.9% | 61.9% |
| LMCA | 47% | 15.9% |
| Epoch Capabilities Index | 154.9 | 132.2 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 27.7% |
| Kagi LLM Benchmark | — | 55.9% |
| ARC-AGI-1 | — | 4.4% |
| EnigmaEval | — | 0.6% |
| Mystery Game Puzzles | 43% | — |
| Surface Evolver Bench | 46.3% | — |
| ForecastBench | — | 57.5 |

## Math

- DeepSeek V4.1 Flash: 66.7 (#25)
- Llama 4 Maverick: 26.0 (#262)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 20.6% |
| LMArena Math | 1477 | 1299 |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 26.8% | — |
| ProofBench | 54% | — |
| Omni-MATH | — | 42.2% |
| MATH Level 5 | — | 73% |
| FrontierMath (Feb 2025 set) | — | 0.7% |

## Knowledge

- DeepSeek V4.1 Flash: 57.9 (#38)
- Llama 4 Maverick: 33.4 (#204)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 89.8% | 67% |
| LMArena Expert | 1506 | 1259 |
| Humanity's Last Exam | — | 5.7% |
| MMLU-Pro | — | 81% |
| Confabulations | — | 22.6% |
| Vectara Hallucination Rate | — | 8.2% |
| GPQA (HELM) | — | 65% |

## Multimodal

- DeepSeek V4.1 Flash: 39.1 (#61)
- Llama 4 Maverick: 31.6 (#105)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 1277 | 1142 |
| GeoBench | — | 52% |
| Furniture Assembly | 34.2% | — |
| SpatialViz-Bench | — | 31.8% |

## Multilingual

- DeepSeek V4.1 Flash: 55.0 (#35)
- Llama 4 Maverick: 42.2 (#195)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1448 | 1269 |
| LMArena Chinese | 1497 | 1277 |
| LMArena French | 1452 | 1259 |
| LMArena German | 1484 | 1291 |
| LMArena Japanese | 1412 | 1207 |
| LMArena Korean | 1452 | 1203 |
| LMArena Russian | 1471 | 1286 |
| LMArena Spanish | 1459 | 1293 |

## Instruction Following

- DeepSeek V4.1 Flash: 77.3 (#26)
- Llama 4 Maverick: 71.7 (#146)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1474 | 1267 |
| IFEval | — | 90.8% |

## Long Context

- DeepSeek V4.1 Flash: 45.2 (#47)
- Llama 4 Maverick: 31.4 (#279)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1475 | 1280 |
| Fiction.LiveBench | — | 46.2% |

## Writing & Preference

- DeepSeek V4.1 Flash: 65.4 (#48)
- Llama 4 Maverick: 38.8 (#252)

| Benchmark | DeepSeek V4.1 Flash | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1462 | 1287 |
| LMArena Creative Writing | 1435 | 1267 |
| EQ-Bench Creative Writing | 1540 | 860 |
| LMArena Multi-Turn | 1457 | 1289 |
| Short-Story Creative Writing | — | 62% |
| WildBench | — | 80% |

## FAQ

### Is DeepSeek V4.1 Flash better than Llama 4 Maverick?

DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 30.9 on the Noometry Index.

### Which is cheaper, DeepSeek V4.1 Flash or Llama 4 Maverick?

DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Llama 4 Maverick lists at $0.19 and $0.65.

### Is DeepSeek V4.1 Flash or Llama 4 Maverick better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 26.6 in the Noometry coding category.

### Which has the bigger context window?

DeepSeek V4.1 Flash does, with 1M tokens against 128K.

### How many benchmarks do DeepSeek V4.1 Flash and Llama 4 Maverick share?

28 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Llama 4 Maverick has 54.