Skip to content

Model Library

Nous ResearchReleased August 27, 2025

Hermes 4 405B

Nous Research · Released August 27, 2025

View public research Browse the library

Intelligence score

9

At a glance

Intelligence
9
Speed
37.2tokens/sec
Latency
2.39stime to first token
Input price
$1.00per 1M tokens
Output price
$3.00per 1M tokens
Context window
130Ktokens
Parameters
406billions

Summary

Comparison summary

Hermes 4 - Llama-3.1 405B (Non-reasoning) scores 9 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 17). Hermes 4 - Llama-3.1 405B (Non-reasoning) generates output at 37.2 tokens per second (based on the median across providers serving the model), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 58.3 t/s). Hermes 4 - Llama-3.1 405B (Non-reasoning) costs $1.00 per 1M input tokens (somewhat higher than average, median: $0.60) and $3.00 per 1M output tokens (somewhat higher than average, median: $2.40), based on the median across providers serving the model.

Specifications

Technical specifications

Model nameHermes 4 405B
ProviderNous Research
Release dateAugust 27, 2025
Intelligence score9
Speed37.2 tokens/sec
Latency (TTFT)2.39s
Input price$1.00 / 1M tokens
Output price$3.00 / 1M tokens
Context window130,000 tokens
Parameters406B
ReasoningNo
Input modalitiestext
Output modalities—
Open weightsYes
LicenseMETA LLAMA 3 COMMUNITY LICENSE AGREEMENT
API availability—
SourceModel Intelligence legacy profile: hermes-4-llama-3-1-405b

Pricing

Token prices

Price per 1M tokens

Hermes 4 405B, as listed in the catalog record

Input $1.00 and output $3.00 per 1M tokens. Missing prices are hatched and marked "Not published".
  1. Input$1.00
  2. Output$3.00

Model Intelligence legacy profile: hermes-4-llama-3-1-405b

FAQ · 19

Frequently asked questions

Hermes 4 - Llama-3.1 405B (Non-reasoning) was released on August 27, 2025.

Hermes 4 - Llama-3.1 405B (Non-reasoning) was created by Nous Research.

Hermes 4 - Llama-3.1 405B (Non-reasoning) scores 9 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 17).

Hermes 4 - Llama-3.1 405B (Non-reasoning) generates output at 37.2 tokens per second (based on the median across providers serving the model), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 58.3 t/s).

Hermes 4 - Llama-3.1 405B (Non-reasoning) has a time to first token (TTFT) of 2.39s (based on the median across providers serving the model), which is somewhat higher than average compared to other open weight non-reasoning models of similar size (median: 2.33s).

Hermes 4 - Llama-3.1 405B (Non-reasoning) costs $1.00 per 1M input tokens (somewhat higher than average, median: $0.60) and $3.00 per 1M output tokens (somewhat higher than average, median: $2.40), based on the median across providers serving the model.

Hermes 4 - Llama-3.1 405B (Non-reasoning) costs $1.00 per 1M input tokens and $3.00 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $1.20 per 1M tokens. Pricing may vary by provider.

No, Hermes 4 - Llama-3.1 405B (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.

Hermes 4 - Llama-3.1 405B (Non-reasoning) supports text input.

Hermes 4 - Llama-3.1 405B (Non-reasoning) supports text output.

No, Hermes 4 - Llama-3.1 405B (Non-reasoning) does not support image input. It can only process text.

No, Hermes 4 - Llama-3.1 405B (Non-reasoning) is not multimodal. It only supports text input.

Hermes 4 - Llama-3.1 405B (Non-reasoning) has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Hermes 4 - Llama-3.1 405B (Non-reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Hermes 4 - Llama-3.1 405B (Non-reasoning) has 406 billion parameters.

Hermes 4 - Llama-3.1 405B (Non-reasoning) is released under the META LLAMA 3 COMMUNITY LICENSE AGREEMENT license. This license allows commercial use.

Hermes 4 - Llama-3.1 405B (Non-reasoning) achieves a score of 9 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Hermes 4 - Llama-3.1 405B (Non-reasoning) is available via API through 1 provider.

Hermes 4 - Llama-3.1 405B (Non-reasoning) is available through 1 API provider.