Nous Hermes 3 Mistral 35B
NEW HOTNous Research · Mistral
Open SourceAPI
A community-driven, instruction-tuned model built on Mistral architecture, optimized for superior chat and reasoning performance.
TextCodeReasoning
Scores 78% on MMLU, competitive across 4 standard benchmarks.
Data aggregated and editorially reviewed by TrendMing.
Benchmark Performance
MMLU78%
HumanEval74%
MATH53%
GPQA37%
Scores sourced from published evaluations. Higher = better.
Quick Specs
Context Window32K tokens
Parameters35B
Input pricingN/A (self-hosted)
Output pricingN/A (self-hosted)
Released2026-02-15