Chat Models — GPT, Claude, Gemini & More
Every leading AI chat model in one place Krater gives you 500+ AI models from OpenAI, Anthropic, Google, Meta, DeepSeek, xAI, Mistral, Moonshot AI, Qwen and dozens more providers, all inside one ch...
Every leading AI chat model in one place
Krater gives you 500+ AI models from OpenAI, Anthropic, Google, Meta, DeepSeek, xAI, Mistral, Moonshot AI, Qwen and dozens more providers, all inside one chat. Pick a model yourself or let an Auto Router choose for you. Browse the full catalog in the models directory.
Featured models
New releases land in the picker as providers ship them. Current highlights:
| Provider | Models |
|---|---|
| OpenAI | GPT-6 Astra, GPT-6 Luna / Sol, GPT-5.6 Luna / Sol / Terra, GPT-5.5 |
| Anthropic | Claude Opus 5.5, Claude Sonnet 5.5, Claude Fable 5.1 |
| Gemini 3.5 to 3.8 Flash, Gemini 3.1 Pro | |
| DeepSeek | DeepSeek V4 Pro and Flash |
| Moonshot AI | Kimi K3 |
| MiniMax | MiniMax M3 |
| Mistral | Mistral Large 4 |
| xAI | Grok 4 series |
| Meta | Llama 4 Maverick, Llama 3.3 |
| Alibaba | Qwen3.7, Qwen3.8 |
Plus hundreds more frontier, open-weight and specialist models, all searchable in the picker.
Auto Routers
Don’t want to choose? Pick a cost tier. The five Auto Routers send each prompt to a model suited to that tier:
| Router | What it picks | Picker tier |
|---|---|---|
| Lite Auto Router | Krater picks the most cost-efficient model for the job | $ |
| Smart Auto Router | Krater picks the best model for every prompt | $$ |
| Power Auto Router | Krater picks a stronger model for harder work | $$$ |
| Elite Auto Router | Krater picks a top-tier model for demanding tasks | $$$ |
| Frontier Auto Router | Krater picks the most capable models, regardless of price | $$$ |
Messages are billed at the picked model’s real cost. Lite costs least, and Frontier costs most. Auto Routers require an Ultra or Max plan. The exception: persona chats use the Smart router on every plan.
Switching models
- Click the model selector in the chat input.
- Search by model or provider, sort by popularity, newest, name or cost, or use the quick filters (Fast, Vision, Private, Cheap).
- Pick a model. Your conversation context carries over.
The picker shows each model’s estimated cost per message and a cost tier badge: Budget under 0.21 credits, Standard 0.21 to 1.5, Pro 1.5 and up.
What a message costs
Chat bills on actual token usage — your input plus the model’s reply. Minimum charge is 0.2 credits per message and a single message can cost at most 500 credits. The estimate next to the send button is a preview; the exact cost shows on the reply. Details: Understanding Credits.
Context window by plan
How much conversation and attached files a model can see at once depends on your plan:
| Plan | Context window |
|---|---|
| Pro | 150,000 tokens |
| Ultra | 400,000 tokens |
| Max | 1,000,000 tokens |
Which model to use?
| Use case | Try |
|---|---|
| Everyday questions | Gemini Flash, GPT-5.6 Luna, DeepSeek V4 |
| Complex reasoning | Power Auto Router, GPT-5.5 Pro, Claude Opus 5.5 |
| Writing and editing | Claude Sonnet 5.5, Claude Fable 5.1 |
| Coding | Power Auto Router, DeepSeek V4, GPT-6 |
| Understanding images | GPT-5.6, Gemini Flash |
| Huge documents | Max plan, 1M-token context models |
Run the same prompt through several models at once with /compare, or let Krater Agent pick tools and models for the whole task.