Hermes 3 70B Instruct
Nous Research · released Jul 29, 2024 · NousResearch/Hermes-3-Llama-3.1-70B
- Type
- Open weightsLlama 3 Community License
- Params
- 70.6B
- Context
- 131K
about 98K words of context · download allowed, licence restricts use
Our take
Written Aug 3, 2026Hermes 3 is a 70.6-billion-parameter text-only model from Nous Research with a 131,072-token request limit and a restricted community licence. It follows instructions well but trails similarly sized models on raw knowledge and reasoning tests.
Pick this for long-context text tasks where downloadable weights matter, or for instruction-following workflows where format compliance is more important than encyclopaedic knowledge. It is a viable budget option when uniform pricing across hosts simplifies cost planning. Skip it if you need unrestricted commercial use, image or audio input, or strong reasoning and knowledge performance from a 70-billion-parameter model.
The case for it
- Strong instruction-following relative to its knowledge benchmarks: IFEval 76.6% sits well above its MMLU-Pro 41.4%, showing format compliance outpaces raw knowledge.
- Very long request limit for a downloadable model of this size, with the full 70.6 billion parameters attending across all 131,072 tokens.
- Uniform pricing across the two cheapest tracked hosts, with no provider markup variation.
The case against it
- Weak raw knowledge and reasoning for its parameter class, with MMLU-Pro 41.4% and GPQA Diamond 14.9% both below typical 70-billion-parameter results.
- Only one of three tracked offers discloses throughput, leaving most hosting performance unverified.
- Restricted licence requires a separate commercial agreement, unlike permissive Apache or MIT alternatives.
How good is it?
IntelligencePuzzles, maths, exam questions
Hermes 3 70B Instruct is not on Arena Text (overall), which is where the rating would come from, so there is no rating here. It is on GPQA Diamond, in 5th of 16 with 14.9.
CodingWriting and fixing code on its own
Nobody we watch has scored Hermes 3 70B Instruct for this. We would take the rating from Arena Coding.
AgenticPlanning, calling tools, staying on task
Nobody we watch has scored Hermes 3 70B Instruct for this. We would take the rating from Arena Agent (IPS).
WritingWe do not rate this
Nobody we watch has scored this model for writing. Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Can you run it yourself?
- Fits in memory
- weights load entirely on the card
- Spills to system RAM
- some weights offload; much slower
- Too large
- will not load even with offload
- est
- size is calculated; the verdict could change by 10%
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Max (32-core GPU) · 64 GB
Borderline fit on an estimated size. It leaves 0.3 GB spare on a size we calculated rather than measured, and a 10% error either way would change the answer.
Memory use by level
Against a 24 GB card.
All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.
Check against your own machine → · All 71 devices, with every size →
Or rent it from someone else
Cheapest of 3 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.
- per 1M tokens
- $0.70 in / $0.70 out
- Context served
- 131K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $0.70 / $0.70 | 131K | not measured | Unknown | Unknown | Unknown |
| DeepInfrafp8 | $0.70 / $0.70 | 131K | not measured | Unknown | Unknown | Unknown |
| DeepInfrafp8 | $0.70 / $0.70 | 131K | 33 tok/s | No | No | Confirmed |
DeepInfraAcross the 3 listings we hold: 1 say they do not train on prompts, 0 say they do and 2 do not say. 1 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouter | ✗ | ✓ | ✓ |
| DeepInfrafp8 | |||
| DeepInfrafp8 | ✗ | ✓ | ✓ |
Tool calling: 0 of 3 listings say yes, 2 say no, 1 publishes no parameter list. JSON output: 2 of 3 listings say yes, 1 publishes no parameter list. Strict schema: 2 of 3 listings say yes, 1 publishes no parameter list.
Models people weigh against Hermes 3 70B Instruct
When we formed this view
Dates behind this page
Prices last checked 35h ago
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- 1 of 3 listings publish no parameter list, so what their API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 2 of 3 listings do not say whether they train on prompts.
- We hold no cached-input rate for any of its listings.
Licence and identifiers
What the licence allowsLlama 3 Community License, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Llama 3 Community License
Commercial use allowed below 700M MAU; requires "Built with Meta Llama 3" attribution and Llama naming on derivatives.
Identifiers
- Hugging Face
- NousResearch/Hermes-3-Llama-3.1-70B
- Architecture
- Dense
- Modality record
- text->text
- Catalogue slug
- nousresearch-hermes-3-70b-instruct