Specification
most-agreed values
Context
8K
Max output
8K
Released
—
Knowledge cutoff
—
Retires
—
Open weights
—
Input
—
Output
—
Quality
3 benchmarks · 1 source
Composite
31st
percentile of 334 scored models
Rank
232 / 334
±0.06 sd
Evidence
3 × 1
one source, or fewer than 3 benchmarks
Effort range
—
one configuration only
reasoning34th1/4 bench
math27th2/7 bench
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| MATH level 5 | 49.8 ±1.2 | default | Epoch AI | 2025-01-27 |
| GPQA diamond | 50.9 ±2.6 | default | Epoch AI | 2025-01-27 |
| OTIS Mock AIME 2024-2025 | 9.7 ±3.2 | default | Epoch AI | 2025-02-25 |
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 1 host
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| snowflake | — | — | — | — | 8K | 8K | — |
Other listings of this model
this page is one of them
The corpus files llama-3-1-405b under several keys. This page is the llama-3-1-405b listing. The full record — every host, every price and every benchmark score — is on the main llama-3-1-405b page.
More from Meta
most-hosted first