Hermes 4 Large
Advanced reasoning model built on Llama-3.1-405B with hybrid thinking modes. Features internal deliberation capabilities, excels at math, code, STEM, and logical reasoning while supporting structured outputs with improved steerability and neutral alignment.
- Structured Output
Added Aug 26, 2025
Model weightsPricing
Auto routing · per 1M tokens- Input
- $0.30
- Output
- $1.20
Specifications
- Context window
- 128K
- Max output
- 8.2K
- Parameters
- 405B
- Avg output (7d)
- 279 tokens
- Longer than 24% of models
Benchmarks
Benchmarks
Sourced from Artificial Analysis.
Intelligence Index
7.4
Agentic work
T²-Bench Telecom (legacy)
Legacy fallback · Conversational AI agents in dual-control scenarios
26.6%
Better than 32% of models compared
Document reasoning
AA-LCR v1.1
Long context reasoning with updated grading
22.0%
Better than 24% of models compared
Reasoning
HLE
Humanity's Last Exam
4.2%
Better than 16% of models compared
IFBench
Instruction-following benchmark
34.8%
Better than 26% of models compared
Coding
Terminal-Bench Hard (legacy)
Legacy fallback · Agentic coding and terminal use
9.8%
Better than 44% of models compared
LiveCodeBench
Contamination-free coding benchmark
54.6%
Better than 62% of models compared
Math
AIME 2025
American Invitational Mathematics Examination 2025
15.3%
Better than 19% of models compared
Knowledge
MMLU-Pro
Professional and academic subject knowledge
72.9%
Better than 42% of models compared
Legacy benchmarks
GPQA Diamond (legacy)
Graduate-level scientific reasoning
53.6%
Better than 28% of models compared
AA-LCR (unversioned / legacy)
Long context reasoning evaluation
22.0%
Better than 24% of models compared
Last updated Oct 3, 2026
Artificial AnalysisProviders
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…