Llama 4 Maverick
| Released | 2025-04-05 |
|---|---|
| Price per 1M | Not listed |
| Overall score | 36.9 · rank 207 |
| Provider | Meta |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 36.9 | 207 | 21 |
| Overall: Coding | 34.4 | 213 | 11 |
| Overall: Reasoning | 40.0 | 193 | 12 |
| Overall: Agentic work | 37.0 | 206 | 11 |
| Overall: Frontend | 34.0 | 207 | 4 |
| Coding | 31.9 | 201 | 2 |
| Agents | 38.0 | 165 | 1 |
| Reasoning | 41.4 | 179 | 8 |
| Math | 38.5 | 169 | 3 |
| Knowledge | 39.3 | 141 | 1 |
| General | 35.0 | 175 | 4 |
| Multimodal | 40.5 | 104 | 2 |
Evidence
Coding 31.9 ±1.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Aider Polyglot | 15.6% | Epoch AI | 2026-11-11 |
| SciCode | 31.7% | Epoch AI | 2026-11-11 |
Agents 38.0 ±3.6
Reasoning 41.4 ±1.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 0.0% | Epoch AI | 2026-11-11 |
| Extended NYT Connections | 8.0% | Lech Mazur | 2026-11-10 |
| ARC-AGI-2 (ARC Prize verified) | 0.0% | ARC Prize | 2026-11-11 |
| GPQA Diamond | 67.0% | Epoch AI | 2025-04-08 |
| CritPt | 0.0% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 4.4% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 4.4% | Epoch AI | 2026-11-11 |
| EnigmaEval | 0.6% | Epoch AI | 2026-11-11 |
| Kagi Reasoning Benchmark | 55.9% | Kagi | 2026-11-11 |
| SimpleBench | 27.7% | SimpleBench | 2026-11-11 |
| SimpleBench | 27.7% | Epoch AI | 2026-11-11 |
Math 38.5 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 (2025 set) | 0.7% | Epoch AI | 2025-04-08 |
| MATH Level 5 | 73.0% | Epoch AI | 2025-04-08 |
| OTIS Mock AIME 2024-2025 | 20.6% | Epoch AI | 2025-04-08 |
Knowledge 39.3 ±1.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Humanity's Last Exam | 5.7% | Epoch AI | 2026-11-11 |
General 35.0 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LLM Debate Benchmark | 1,021 | Lech Mazur | 2026-10-26 |
| EQ-Bench Creative Writing v3 | 858 | EQ-Bench | 2026-11-11 |
| LMArena Text | 1,327 | LMArena | 2026-11-09 |
| EQ-Bench Longform Creative Writing | 31.9% | EQ-Bench | 2026-11-11 |
Multimodal 40.5 ±4.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Vision | 1,145 | LMArena | 2026-11-09 |
| GeoBench | 52.0% | Epoch AI | 2026-11-11 |
Names matched to this model
llama-4-maverick, llama-4-maverick-17b-128e, llama-4-maverick-17b-128e-instruct, llama-4-maverick-17b-128e-instruct-fp8