Hermes 4 405B

NousResearch · released 2025-08-26 · llama3 license

Hermes 4 405B is Nous Research's hybrid-reasoning fine-tune of Llama-3.1-405B (Aug 2025), with technical-report reasoning-mode scores of MATH-500 96.2, AIME25 78.1, GPQA 70.6 and LiveCodeBench 61.4, plus class-leading low refusal rates.

Key specs

TypeLocal open-weight
Parameters405.85B total
Architecturellama
Context window131K tokens
Knowledge cutoff2024-08-31
Modalitiestext
Recommended backends
Minimum viable rig

Benchmark scores

GPQA Diamond70.6%
SWE-bench Verified
AIME78.1%
MMLU-Pro80.6%
BFCL v3 (tool use)
Composite score7.66
Community ratingNo reviews yet

VRAM & disk per quantization

QuantVRAMDiskRAMContext
Q4_K_M236.9 GB235.4 GB131K

Strengths & weaknesses

Strengths: Frontier-level math (MATH-500 96.2, AIME25 78.1) in reasoning mode; Very low refusal / highly steerable; Hybrid reasoning with explicit think mode

Weaknesses: Built on Llama-3.1-405B; non-reasoning mode scores drop sharply; Large/expensive to self-host; restrictive Llama license