> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mantle.chat/llms.txt
> Use this file to discover all available pages before exploring further.
# Multiple AI Models
> Access GPT, Claude, Gemini, Grok, DeepSeek and more from one platform
Mantle Chat gives you access to the world's leading AI models — no need to manage separate subscriptions.
## Available Models
### Anthropic (Claude)
| Model | Cost | Best For | Context |
| --------------------- | -------- | -------------------------------------- | ------- |
| **Claude Opus 4.6** | \$\$\$\$ | Frontier reasoning, agentic mastery | 1M |
| **Claude Opus 4.5** | \$\$\$\$ | Deep analysis, complex problem solving | 200K |
| **Claude Sonnet 4.6** | \$\$\$\$ | Advanced coding, precise instructions | 1M |
| **Claude Sonnet 4.5** | \$\$\$\$ | Complex agents, advanced coding | 200K |
| **Claude Sonnet 4** | \$\$\$\$ | Superior coding, precise instructions | 200K |
| **Claude Haiku 4.5** | \$\$\$ | Real-time apps, fast responses | 200K |
### OpenAI (GPT)
| Model | Cost | Best For | Context |
| ---------------- | ------ | ------------------------------------- | ------- |
| **GPT-5.4** | \$\$\$ | Complex reasoning, advanced coding | 1M |
| **GPT-5.4 Mini** | \$\$\$ | Balanced performance, cost efficiency | 200K |
| **GPT-5.4 Nano** | \$\$ | Ultra-low latency, high concurrency | 128K |
| **GPT-5.2** | \$\$\$ | Advanced coding, complex analysis | 500K |
| **GPT-5.1** | \$\$\$ | Fast coding, low-latency tasks | 400K |
| **GPT-5** | \$\$\$ | High-speed responses, multimodal | 128K |
| **GPT-5 Mini** | \$\$ | Balanced performance, cost efficiency | 128K |
| **GPT-5 Nano** | \$ | Ultra-low latency, high concurrency | 64K |
| **GPT-4.1** | \$\$\$ | Complex problem solving, engineering | 1M |
| **GPT-4.1 Mini** | \$\$ | Balanced performance, efficient | 128K |
| **GPT-4.1 Nano** | \$ | Ultra-fast responses, real-time apps | 64K |
| **GPT-4o** | \$\$\$ | Multimodal tasks, complex reasoning | 128K |
| **GPT-4o Mini** | \$ | Fast processing, cost-effective | 128K |
### OpenAI (o-series Reasoning)
| Model | Cost | Best For | Context |
| ----------- | ------ | --------------------------------------- | ------- |
| **o3** | \$\$\$ | Complex reasoning, math, science | 200K |
| **o3 Mini** | \$\$\$ | Advanced reasoning, scientific analysis | 128K |
| **o4 Mini** | \$\$\$ | Mathematical reasoning, logic | 128K |
### Google (Gemini)
| Model | Cost | Best For | Context |
| ------------------------- | ------ | ------------------------------------- | ------- |
| **Gemini 3.1 Pro** | \$\$\$ | Advanced reasoning, agentic workflows | 1M |
| **Gemini 3 Flash** | \$\$ | Fast processing, large scale tasks | 1M |
| **Gemini 3.1 Flash-Lite** | \$\$ | Cost-efficiency, high throughput | 1M |
| **Gemini 2.5 Pro** | \$\$\$ | Coding, reasoning, multimodal | 1M |
| **Gemini 2.5 Flash** | \$\$ | Large scale processing, agentic | 1M |
| **Gemini 2.5 Flash Lite** | \$ | Cost-efficiency, high throughput | 1M |
### xAI (Grok)
| Model | Cost | Best For | Context |
| ----------------- | ------ | -------------------------------------- | ------- |
| **Grok 4** | \$\$\$ | Natural language, math, reasoning | 256K |
| **Grok 4.1 Fast** | \$ | Agentic tool calling, high performance | 2M |
### DeepSeek
| Model | Cost | Best For | Context |
| ----------------- | ---- | ------------------------------------ | ------- |
| **DeepSeek V3.2** | \$ | Math reasoning, coding, writing | 128K |
| **DeepSeek R1** | \$ | Agentic reasoning, structured output | 128K |
***
## How to Switch Models
Find it at the bottom of your chat window
Browse by provider or search by name (`Cmd/Ctrl + K`)
Your conversation context is preserved
Use **Cmd/Ctrl + K** to quickly search and switch models without leaving your keyboard.
***
## Model Selection Tips
**Claude Sonnet 4.6**, **Claude Sonnet 4.5**, or **Gemini 2.5 Pro** — excellent at code generation, debugging, and
complex refactoring. For cost-effective coding, try **GPT-4.1**.
**Claude Sonnet 4** or **Claude Sonnet 4.5** — natural writing style, great at following tone and format
instructions.
**o3**, **Claude Opus 4.6**, or **Grok 4** — when you need step-by-step thinking on hard problems. For math
specifically, try **DeepSeek R1**.
**Claude Haiku 4.5**, **GPT-4.1 Nano**, or **Gemini 2.5 Flash Lite** — fast and cost-effective for simple questions.
**Gemini 2.5 Pro** (1M context) or **Grok 4.1 Fast** (2M context) — when you need to analyze large files or
codebases.
For detailed model benchmarks and comparisons, check out [Artificial Analysis](https://artificialanalysis.ai/) — an
independent resource for AI model performance data.
***
## Special Capabilities
### Web Search
Some models can search the web for current information:
* Returns results with **citations and sources**
* Great for research, news, and fact-checking
* Available on Pro and Team
### Reasoning Mode
See the AI's thinking process:
* Models with reasoning support show chain-of-thought thinking
* Helpful for understanding complex answers
* Toggle visibility in settings
### File Analysis
Upload files for AI to analyze:
* Documents (PDF, Word, text)
* Images (for multimodal models)
* Code files
***
## Model Access by Plan
| Plan | Model Tiers | Example Models |
| -------- | -------------------------- | ------------------------------------------------------------------------------------ |
| **Free** | Curated access | GPT-4.1 Nano, GPT-5 Nano, GPT-4o Mini, GPT-4.1 Mini, Gemini 2.5 Flash, Grok 4.1 Fast |
| **Pro** | \$, \$\$, \$\$\$, \$\$\$\$ | All models including Claude Sonnet 4.6 and Claude Opus 4.6 |
| **Team** | \$, \$\$, \$\$\$, \$\$\$\$ | Same full model access as Pro inside org workspaces |
See full plan comparison and pricing
***
## Resources
Independent AI model benchmarks and performance comparisons
Official OpenAI documentation
Official Claude model documentation
Official Gemini model documentation