Nous Hermes vs Llama vs Mistral AI vs Zephyr
A side-by-side look at scores, pricing and features — with RECATOOLS' ASEAN-aware verdict for each.
Nous Hermes
Nous Research's steerable open-weight models, now Hermes 4 and 4.3
Visit
|
Llama
Meta's downloadable LLM family, last updated with Llama 4 in April 202...
Visit
|
Mistral AI
Paris model lab with open weights, cheap APIs and the Vibe agent
Visit
|
Zephyr
Hugging Face's 2023 DPO-tuned Mistral 7B, now a reference model
Visit
|
|
|---|---|---|---|---|
| RECATOOLS Score | 6.8 / 10 | 8.3 / 10 | 8.4 / 10 | 5.8 / 10 |
| Capability | ||||
| Value for money | ||||
| Ease of use | ||||
| ASEAN readiness | ||||
| API quality | ||||
| Pricing | Open Source | Open Source | Freemium | Open Source |
| Free tier | Open weights free to download and run, under Apache 2.0 or the Llama 3 licence depending on the model | Free weight downloads (Llama 2, 3.x and 4) under Meta's Llama community licences, after accepting the licence | Vibe Free: limited messages, web searches and coding sessions, basic image generation, 100+ connectors; Apache 2.0 weights free to self-host | MIT-licensed weights, ungated and free to download; self-hosted, or through a third-party inference provider at its rates |
| Paid from | Free weights, or hosted Hermes 4 405B on OpenRouter from $1 per million input tokens | — | $14.99/month (Vibe Pro, excluding tax); API from $0.10 per million tokens | — |
| Has API | ✓ | ✓ | ✓ | ✓ |
| Open source | ✓ | ✓ | ✓ | ✓ |
| Free to use | ✓ | ✓ | ✓ | ✓ |
| Users | 500k+ downloads | 10m+ downloads | 500k+ API users | 500k+ downloads |
| Founded | 2023 | 2023 | 2023 | 2023 |
| Maker | Nous Research, Inc. | Meta Platforms | Arthur Mensch, Guillaume Lample, Timothee Lacroix | Hugging Face |
| Verdict | Hermes remains the open-weight line to reach for when you want a model that follows your own system prompt: Hermes 4 brought hybrid reasoning in 14B, 70B and 405B sizes, and Hermes 4.3 squeezes similar quality into 36B w... |
Llama 2 and 3 made Meta's family the default starting point for self-hosted and fine-tuned models, and the weights from Llama 2 through Llama 4 remain free to download from Hugging Face and Kaggle. Llama 4 Scout and Mave... |
Mistral is a good choice when you want models you can host yourself or buy cheaply by the token: Small 4, Large 3 and the Ministral 3 models are Apache 2.0, and API rates sit well below most US frontier labs. The Vibe ag... |
Zephyr-7B-β earned its place in October 2023: a 7B model whose MT-Bench score beat Llama2-Chat-70B, with the whole distilled-DPO recipe published in the alignment-handbook. The MIT-licensed weights are still ungated and... |
| Full review → | Full review → | Full review → | Full review → |
Comparisons cover up to 4 tools. Scores are RECATOOLS editorial assessments; verify current pricing on each vendor's site.