Skip to content
Alpha
  • Models
  • Rankings
  • Ori

Footer

OpenRouterOpenRouter
© 2026 OpenRouter, Inc

Product

  • Chat
  • Rankings
  • Benchmarks
  • Apps
  • Discover
  • Models
  • Ori
  • Collections
  • Providers
  • Tools
  • Pricing
  • Business
  • Enterprise
  • Labs

Company

  • About
  • Blog
  • Customers
  • Careers
    Hiring
  • Privacy
  • Terms of Service
  • Trust Center
  • Support
  • Works With OR
  • Data
  • Brand

Developer

  • Documentation
  • API Reference
  • Developer Platform
  • Status
  • AI Site Map

Connect

  • Discord
  • GitHub
  • LinkedIn
  • X
  • YouTube
Favicon for nousresearch

Nous: Hermes 4 405B

nousresearch/hermes-4-405b

Model weights
Compare

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 <think>...</think> 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 docsOpens in new tab

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.

Modalities
In / Out Price
$1 / $3per 1M
Context
131K
Released
Aug 26, 2025
Knowledge Cutoff
Aug 2024
Compare
ProvidersPricingPerformanceUptimeBenchmarksAppsActivityFAQExplore

Providers

This model is hosted by one provider. OpenRouter forwards every request to it directly — no routing decisions to make.

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Nous: Hermes 4 405B (Artificial Analysis)
SourceBenchmarkScore
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) Math Index15.3
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) GPQA Diamond53.6%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) HLE4.2%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) IFBench34.8%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) τ²-Bench Telecom26.6%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) AA-LCR22.0%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) CritPt0.0%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) LiveCodeBench54.6%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) Terminal-Bench Hard9.8%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) MMLU-Pro72.9%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Non-reasoning) AIME 202515.3%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) Math Index69.7
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) GPQA Diamond72.7%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) HLE10.9%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) IFBench32.7%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) τ²-Bench Telecom22.2%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) AA-LCR22.3%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) CritPt0.3%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) LiveCodeBench68.6%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) Terminal-Bench Hard11.4%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) MMLU-Pro82.9%
Artificial AnalysisHermes 4 - Llama-3.1 405B (Reasoning) AIME 202569.7%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Not currently available through EU in-region routing. In-region routing docs

Explore more models

AI Model RankingsRanking
Latency / throughput
$1.00$3.00$0.200.40s21 tps
100.00%
Throughput
21tok/s

P50, best across providers

Latency
0.40s

P50, best provider

Uptime (3d)The model was reachable. Request routed to a provider.
100.00%
Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.
99.96%

Availability over the last 3 days

Last 72 hours
Availability 99.96%
3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability
99.91%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

1.
Favicon for https://sillytavern.app/
SillyTavern
SillyTavern is the LLM frontend for power users, a chat interface that connects to any model and gives deep control through character creation, roleplay, and prompt customization.
155Mtokens
2.
Favicon for https://creatoros.app/
CreatorOS Mentor
new
62.5Mtokens
3.
Favicon for https://godmod3.ai/
G0DM0D3-pliny
new
42.5Mtokens
4.
Favicon for https://art-from-the-machine.github.io/
Mantella
Skyrim & Fallout 4 mod, naturally speak to NPCs
29.9Mtokens
5.
Favicon for https://mia.local/
Auresys - Multimodal Intelligence Agent
new
25.8Mtokens

Frequently asked questions

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 <think>...</think> traces or respond directly, offering flexibility between speed and depth. Users can control the reasoning behaviour with the reasoning enabled boolean.

Hermes 4 405B costs $1.00/M input tokens and $3.00/M output tokens, with separate rates for Cache Read at $0.20/M tokens.

Hermes 4 405B has a 131,072 token context window.

No. Hermes 4 405B does not accept tools, so function calling is unavailable. It also supports structured outputs via a JSON schema in response_format.

Hermes 3 70B Instruct and Hermes 3 405B Instruct are other text models from Nous Research.

Hermes 4 405B was released on August 26, 2025. Its knowledge cutoff is August 31, 2024.