Hermoso renders through 22 text models on one account and one credit balance. You can name a model, or leave it out and get a sound default — the same catalog is available from the app, the MCP server and the CLI.
Claude Fable 5.1
Most capable — deepest creative + reasoning
Claude Fable 5
Previous Fable
Claude Opus 5
Newest — deepest analysis
Claude Opus 4.8
Best for complex, analytical work
Claude Opus 4.6
Best for long-form creative work
Claude Sonnet 5
Responsive everyday work
Claude Sonnet 4.6
Previous generation
Gemini 3.1 Pro
Google’s deepest reasoning + writing
Gemini 3.5 Flash
Google — fast + sharp
Gemini 3.8 Flash
Newest Google — fast, strong everyday work
Gemini 3.7 Flash
Fast Google — coding and agentic tool use
Gemini 3.6 Flash
Previous-generation Flash
Gemini 3.5 Flash-Lite
Cost-optimized Google workhorse
Gemini 3.1 Flash-Lite
Cheapest current-gen Gemini text
GPT-5
OpenAI’s flagship — broad + reliable
GPT-5 Mini
Fast + cheap OpenAI
GPT-OSS 120B
OpenAI’s open-weight flagship — strong + open
DeepSeek V3.1
Open frontier — strong reasoning, great value
DeepSeek R1
Open reasoning model — deliberate step-by-step
Meta’s open flagship
Meta’s efficient open MoE — fast, long-context
Open workhorse — dependable + great value
Side by side
| Model | Notes |
|---|---|
| Claude Fable 5.1 | Most capable — deepest creative + reasoning |
| Claude Fable 5 | Previous Fable |
| Claude Opus 5 | Newest — deepest analysis |
| Claude Opus 4.8 | Best for complex, analytical work |
| Claude Opus 4.6 | Best for long-form creative work |
| Claude Sonnet 5 | Responsive everyday work |
| Claude Sonnet 4.6 | Previous generation |
| Gemini 3.1 Pro | Google’s deepest reasoning + writing |
| Gemini 3.5 Flash | Google — fast + sharp |
| Gemini 3.8 Flash | Newest Google — fast, strong everyday work |
| Gemini 3.7 Flash | Fast Google — coding and agentic tool use |
| Gemini 3.6 Flash | Previous-generation Flash |
| Gemini 3.5 Flash-Lite | Cost-optimized Google workhorse |
| Gemini 3.1 Flash-Lite | Cheapest current-gen Gemini text |
| GPT-5 | OpenAI’s flagship — broad + reliable |
| GPT-5 Mini | Fast + cheap OpenAI |
| GPT-OSS 120B | OpenAI’s open-weight flagship — strong + open |
| DeepSeek V3.1 | Open frontier — strong reasoning, great value |
| DeepSeek R1 | Open reasoning model — deliberate step-by-step |
| Llama 4 Maverick | Meta’s open flagship |
| Llama 4 Scout | Meta’s efficient open MoE — fast, long-context |
| Llama 3.1 70B | Open workhorse — dependable + great value |
One account and one credit balance covers every model. Each render is priced before it runs and settled at its exact cost, and the current price per model is shown in the app beside the button that spends it — so it is never out of date. See pricing for what a credit is worth, or every model we offer.
Frequently asked questions
How many text models does Hermoso support?
22 as of this build, read from the live catalog rather than a list kept by hand.
Do I need a separate subscription for each model?
No. One account and one credit balance covers every model in the catalog; you are charged the exact credit cost of the render you make.