Skip to main content

Hermes 4 405B

Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with \u003cthink\u003e...\u003c/think\u003e traces or respond directly, offering flexibility between speed and depth. Users can control the reasoning behaviour with the reasoning enabled boolean. Learn more in our docs The model is instruction-tuned with an expanded post-training corpus (~60B tokens) emphasizing reasoning traces, improving performance in math, code, STEM, and logical reasoning, while retaining broad assistant utility. It also supports structured outputs, including JSON mode, schema adherence, function calling, and tool use. Hermes 4 is trained for steerability, lower refusal rates, and alignment toward neutral, user-directed behavior.

Model card on Hugging Face
Input / output price
$1.00 / $3.00 per 1M
Context
131K tokens
Released
Aug 26, 2025
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "hermes-4-405b".

Read docs

Providers

2 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.

ProviderInputOutput
Nebiusnebius/hermes-4-405b
$1.00/M$3.00/M
OpenRouteropenrouter/hermes-4-405b
$1.00/M$3.00/M

Is a provider down? OpenRouter status

Pricing

Averaged across 2 providers, in USD per 1M tokens after gateway discounts.

Average input
$1.00 per 1M
Average output
$3.00 per 1M
Average cache read
$0.20 per 1M

Cost by volume

Total cost from 0 to 1T tokens.

$1.00/MNebius and OpenRouter

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

15tokens/sacross providers

Faster than 30% of the other 599 models on the gateway

Time to first token

Lower is better

0.6sacross providers

Faster than 89% of the other 150 models on the gateway

Response latency

Lower is better

0.7sacross providers

Faster than 77% of the other 780 models on the gateway

Benchmarks

6 published scores from 1 source, each linked to where it was published.

Published scores

BenchmarkScore
LiveCodeBench v6 (Aug 2024+)arxiv.org61.3%

Apps

Use Hermes 4 405B in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

Hermes 4 405B FAQ

Short answers, built from the data on this page.

Hermes 4 405B is a model from Nous Research, released on Aug 26, 2025. Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with \u003cthink\u003e...\u003c/think\u003e traces or respond directly, offering flexibility between speed and depth. Users can control the reasoning behaviour with the reasoning enabled boolean. Learn more in our docs The model is instruction-tuned with an expanded post-training corpus (~60B tokens) emphasizing reasoning traces, improving performance in math, code, STEM, and logical reasoning, while retaining broad assistant utility. It also supports structured outputs, including JSON mode, schema adherence, function calling, and tool use. Hermes 4 is trained for steerability, lower refusal rates, and alignment toward neutral, user-directed behavior.

On the Respan gateway, Hermes 4 405B costs $1.00 per 1M input tokens and $3.00 per 1M output tokens on every provider, after gateway discounts.

Hermes 4 405B is available through 2 providers: Nebius and OpenRouter.

Hermes 4 405B accepts up to 131K tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "hermes-4-405b". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Hermes 4 405B has 6 published benchmark scores, including LiveCodeBench v6 (Aug 2024+) 61.3%, GPQA Diamond 70.5%, and AIME 2024 81.9%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.