| Model | Provider | Status | Context window | Input $/1M tokens | Output $/1M tokens | Quality & price positioning |
|---|---|---|---|---|---|---|
gpt-6-luna |
OpenAI | Available | 1050000 | $0.10 | $0.50 | Default – cheapest OpenAI model, excellent value |
gpt-6.1-sol |
OpenAI | Available | 1050000 | $2.00 | $10.00 | GPT-6.1 – near-Astra quality at a fifth of the price |
gpt-6-astra |
OpenAI | Available | 1050000 | $10.00 | $50.00 | Flagship (GPT-6) – highest quality, very expensive |
claude-haiku-5-5 |
Anthropic | Available | 1000000 | $0.10 | $0.50 | Anthropic – cheapest Claude model, priced like gpt-6-luna (prompts over 100k tokens cost 5×) |
claude-sonnet-5-5 |
Anthropic | Available | 1000000 | $2.00 | $10.00 | Anthropic – very good quality on long context, mid-priced |
claude-opus-5-5 |
Anthropic | Available | 1000000 | $4.00 | $20.00 | Anthropic – highest quality, about twice the price of Sonnet 5.5 |
claude-fable-5-1 |
Anthropic | Available | 1000000 | $10.00 | $50.00 | Anthropic – most capable model, very expensive |
gemini-3.5-flash-lite |
Available | 1048576 | $0.30 | $2.50 | Google – cheapest Gemini model, minimal thinking by default | |
gemini-3.8-flash |
Available | 1048576 | $0.75 | $3.75 | Google – fast, good value (price doubles from 1 Jan 2027) | |
gemini-4-argon |
Available | 1048576 | $2.00 | $10.00 | Google – flagship (Gemini 4), priced like gpt-6.1-sol (doubles from 1 Dec 2026) |
LLM models
Supported AI models
TrustChat talks to every model through a swappable provider adapter – selectable per bot, with no provider-specific code in the domain model.
Prices are the providers' official list prices (USD, as of October 2026) and are meant to guide model choice. The cost report in the super-admin area uses these prices per model. Models marked "Coming soon" were announced recently or still need adapter changes and can't be selected as a bot's model yet. Anthropic and Gemini models become selectable in the admin once that provider's API key is configured.