Meta
Llama 4 Maverick benchmark results
7 results on 7 benchmarks from 3 sources.
Also listed as: llama-4-maverick-17b-128e-instruct · Llama-4-Maverick-17B-128E-Instruct-FP8 · Llama 4 Maverick (FP8) · Llama-4-Maverick-17B-128E-Instruct · Llama-4-Maverick-17B-128E-Instruct-FP8 (FC)
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1327 | 247 | Overall, style control on; 39453 votes; 95% CI 1323–1331“llama-4-maverick-17b-128e-instruct” | Official leaderboard | Oct 8, 2026 |
| LMArena Vision | 1145 | 113 | Overall, style control on; 6923 votes; 95% CI 1136–1154“llama-4-maverick-17b-128e-instruct” | Official leaderboard | Oct 2, 2026 |
| GPQA Diamond | 67.0% | 168 | Epoch AI run, accuracy; effort: FP8 | Independent run | Apr 8, 2025 |
| AIME (OTIS Mock AIME 2024–2025) | 20.6% | 207 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: FP8 | Independent run | Apr 8, 2025 |
| MATH Level 5 | 73.0% | 40 | Epoch AI run, accuracy; effort: FP8 | Independent run | Apr 8, 2025 |
| Humanity's Last Exam | 5.7% | 43 | Compiled by Epoch AI | Third-party compilation via Epoch AI | Apr 5, 2025 |
| BFCL v4 (function calling) | 37.3% | 50 | Overall accuracy; native function calling (FC)“Llama-4-Maverick-17B-128E-Instruct-FP8 (FC)” | Official leaderboard | Oct 17, 2026 |
Related on this site
FAQ
Where do these Llama 4 Maverick scores come from?
From Epoch AI Benchmarking Hub, Berkeley Function Calling Leaderboard (BFCL), and LMArena leaderboard dataset. Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.