NousResearch / Model details

Hermes-4-70B

NousResearch/Hermes-4-70B

fine-tuned

Read the original model card ↗ · Choosing a model

About this model

NousResearch's Hermes-4-70B is a derivative of Llama-3.1-70B, released as safetensors weights under the llama3 license. The model card describes post-training on a newly synthesized corpus emphasizing verified reasoning traces, with the creator claiming improvements in math, code, STEM, logic, creativity, and format-faithful outputs while preserving general assistant quality and broadly neutral alignment. The creator says it uses a hybrid reasoning mode with explicit thinking and response segments, supports function calling and structured JSON outputs, and is easier to steer with reduced refusal rates. According to the model card, Hermes 4 achieves state-of-the-art results on RefusalBench across popular closed and open models in being helpful and conforming to user values without censorship. These capability and benchmark claims have not been independently tested here.

Description sources

https://huggingface.co/NousResearch/Hermes-4-70B/raw/d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09/README.md ↗

Source-based description, not an independent evaluation. Check the linked sources for the creator’s claims and limitations.

Size and formats

Base model: meta-llama/Llama-3.1-70B, meta-llama/Meta-Llama-3.1-70B
Parameters: 70.5537B
Main repository formats: safetensors
License tag: llama3
Languages: en
Last updated: 2025-09-02 · 06:40:36.000 UTC

What changed

Model types are creator-reported, not independently tested. These labels describe source metadata, not a guarantee of behavior.

fine-tunedsource tag: base_model:finetune:meta-llama/Llama-3.1-70B

Family evidence

These groupings are source-reported. Explicit base identifiers take priority over generic architecture tags.

Llama: Base model: meta-llama/Llama-3.1-70B
Llama: Base model: meta-llama/Meta-Llama-3.1-70B

Available versions

Versions appear here only when they explicitly name this exact model as their source. Check each repository for available files.

Main repository files ↗safetensors
No other explicitly linked versions recorded.

Link to this listing

Made or quantized this model? Paste one of these into your model card or README so people can find this listing.

Markdown

[Listed on Custom Clankers](https://customclankers.com/models/nousresearch-hermes-4-70b/)

HTML

<a href="https://customclankers.com/models/nousresearch-hermes-4-70b/">Listed on Custom Clankers</a>

Plain text links only: no image, no script, nothing loaded from this site.

Customize or build on this model

Listed from public GitHub data, not tested by us. Read the code and license before you run anything.

Tools to customize it further

Remove more refusals

Heretic ↗ ★ 36,421
Removes refusal behavior from transformer models automatically. It pairs directional ablation (abliteration) with an Optuna parameter search that tries to cut refusals while keeping the model close to the original, so you don't have to pick layers by hand.
OBLITERATUS ↗ ★ 8,946
A larger refusal-removal toolkit with several abliteration methods, usable locally or through a hosted Hugging Face Space. Check the telemetry setting before you run it: the README describes telemetry as opt-in, but on by default in the Spaces version.

Fine-tune it on your own data

Unsloth ↗ ★ 78,528
A desktop app and Python library for running and training models locally. Its README lists LoRA, QLoRA, full fine-tuning, DPO and GRPO, plus export to formats including GGUF, on Windows, Linux, WSL and macOS.
LlamaFactory ↗ ★ 75,713
Fine-tunes more than 100 language and vision-language models from a command line or a browser-based Web UI, aimed at training without writing code.

Make a smaller download (quantize)

llama.cpp ↗ ★ 132,260
Runs models locally and includes the tools behind most GGUF files. A conversion script turns Hugging Face weights into GGUF, and llama-quantize shrinks a high-precision GGUF to lower-bit formats, which can cost some accuracy.
MLX LM ↗ ★ 7,467
Apple's MLX package for running and fine-tuning models on Apple silicon. It can quantize models and upload them to the Hugging Face Hub, and supports LoRA and full fine-tuning.

Merge it with another model

mergekit ↗ ★ 7,414
Merges pretrained models with many merge methods. It works out-of-core, so merges can run on a CPU or with as little as 8 GB of VRAM, and it also handles LoRA extraction and mixture-of-experts merges.

Learn the techniques

LLM Course ↗ ★ 84,667
A free course with roadmaps and Colab notebooks. Its notebooks include one-click tools for merging with MergeKit, quantizing to GGUF, GPTQ, EXL2, AWQ and HQQ, and automatic abliteration with custom datasets.

Compare all tools →

Source provenance

Downloads: 1,844
Likes: 224
Metadata checked: 2026-10-25 · 12:12:34 UTC
First seen in index: 2026-10-25 · 12:12:34 UTC
Source revision: d5dec2bd6b3930a09ddefd0b7fc6523fe0720d09

Source-card excerpt

Source model card excerpt: Hermes 4 — Llama-3.1 70B Model Description Hermes 4 70B is a frontier, hybrid-mode reasoning model based on Llama-3.1-70B by Nous Research that is

From NousResearch/Hermes-4-70B. Creator model-card excerpt, not an independent review.

Inspect before you run

Low refusal is not proof of correctness. Outputs can be inaccurate, harmful, or biased. Review the model card, license and files yourself. A license tag is not commercial-use advice. Parameter counts are source metadata; no hardware compatibility is implied.

This directory does not host or redistribute model weights. Refusal profiles are measured by us on a fixed prompt set; all other behavior claims are source-reported.

Collection rules & limitations ↗