Tier
Moderate
Headquarters
Distributed
Founded
2023
Flagship
Hermes
Models tracked
7
Artificial Analysis
Top intelligence
18.6
Artificial Analysis Index
Cheapest model
$0.00
blended /1M
Fastest output
91
tokens / sec
Flagship
Nous Research’s top-scoring model on the Artificial Analysis Intelligence Index.
Most intelligent model
Hermes 4 - Llama-3.1 405B (Reasoning)
Released 2025-08-27
Intelligence
18.6
Coding
16.0
Math
69.7
In $/1M
$1.00
Out $/1M
$3.00
Speed t/s
42
Benchmark breakdown
Models & benchmarks
Every Nous Research model on file — intelligence, coding, price, and speed.
| Model | Released | Intel | Coding | Math | In $/1M | Out $/1M | Speed t/s | Latency s |
|---|---|---|---|---|---|---|---|---|
| Hermes 4 - Llama-3.1 405B (Reasoning) | 2025-08-27 | 18.6 | 16.0 | 69.7 | $1.00 | $3.00 | 42 | 0.76 |
| Hermes 4 - Llama-3.1 405B (Non-reasoning) | 2025-08-27 | 17.6 | 18.1 | 15.3 | $1.00 | $3.00 | 39 | 0.76 |
| Hermes 4 - Llama-3.1 70B (Reasoning) | 2025-08-27 | 16.0 | 14.4 | 68.7 | $0.13 | $0.40 | 91 | 0.61 |
| Hermes 4 - Llama-3.1 70B (Non-reasoning) | 2025-08-27 | 12.6 | 9.2 | 11.3 | $0.13 | $0.40 | 88 | 0.64 |
| DeepHermes 3 - Mistral 24B Preview (Non-reasoning) | 2025-03-13 | 10.9 | — | — | $0.00 | $0.00 | 0 | 0.00 |
| Hermes 3 - Llama-3.1 70B | 2024-08-15 | 10.6 | — | — | $0.30 | $0.30 | 32 | 0.37 |
| DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning) | 2025-02-13 | 7.6 | — | — | $0.00 | $0.00 | 0 | 0.00 |
Trajectory
Nous Research’s model releases over time, by intelligence.
House strengths
Where Nous Research’s lineup over-indexes, averaged across its models.
Capability profile
Best value
Hermes 4 - Llama-3.1 70B (Reasoning)
Most intelligence per dollar — 16.0 index at $0.20/1M blended.
81 index points per $1
Overview
An open research collective producing the Hermes line of fine-tuned open models and tooling for decentralized training.
Visit Nous Research