Ctrl K
Back to models

Nous Research

Hermes 4 - Llama-3.1 405B

  • Released
  • Open weights
  • Meta Llama 3 Community License Agreement
Compare
Intelligence Index

Reasoning setting

7.5
Estimated · Artificial Analysis v4.3.2
API price from
$1.00 in / $3.00 out / 1M tokOpenRouter · Nebius · FP8
Context window
128K tok

Performance profile

ReasoningRank #75 of 107
72.7%
GPQA Diamond

Capabilities

Context window

Modalities

Measured setting

Integration

Reasoning trade-offs

Source (opens in a new tab)
Intelligence Index v4.3.2
ReasoningEstimated
Intelligence Index
7.5

Pricing & access

API route
OpenRouter · Nebius · FP8API
Tier & contextStandard

USD per 1M tokens

Input
$1.00
Output
$3.00

Run locally

GGUF
Context
Estimated memory

~253 GiB

RAM + VRAM

Estimate assumptions

One text chat at the selected context, with a 16-bit cache and app allowance. Operating system and image processing are extra; optimized caches can use less.

384 GiB RAMwithout GPU offload

Memory budgetOne text chat
Weights
226.4 GiB
Context cache
3.9 GiB16-bit cache
App allowance
22.6 GiB
Download
243.1 GB7 files · keep together
Disk space
~273 GB
Available software
Ollama (opens in a new tab)LM Studio (opens in a new tab)Jan (opens in a new tab)KoboldCpp (opens in a new tab)llama.cpp (opens in a new tab)
Get GGUF (opens in a new tab)
Set up in Ollama · Q4_K_M · 8K context

Save the following as Modelfile. The context matches the estimate above.

FROM hf.co/lmstudio-community/Hermes-4-405B-GGUF:Q4_K_M
PARAMETER num_ctx 8192

Then run these commands in your terminal:

ollama create llmcurrent-local -f Modelfile
ollama run llmcurrent-local

Ollama import instructions (opens in a new tab) · For LM Studio or Jan, open the app’s model search, enter lmstudio-community/Hermes-4-405B-GGUF, choose Q4_K_M, and set context to 8,192 tokens in the app’s model settings. Software links above open each app’s setup guide.

vLLM setup (opens in a new tab)

Benchmarks

Full leaderboardSource (opens in a new tab)

Humanity’s Last Exam

Frontier reasoning

Hermes 4 - Llama-3.1 405B · Reasoning#89 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 86Granite 4.2 30BReasoning11.2%
  6. 87North Mini CodeReasoning11.1%
  7. 88gpt-oss-20bHigh11.0%
  8. 89Hermes 4 - Llama-3.1 405BReasoning10.9%
  9. 90Nemotron 3.5 LightningReasoning10.6%
  10. 91Qwen3 Coder NextStandard10.1%
  11. 92Qwen3.5 4BReasoning9.9%

Peers use their best measured result on this benchmark.