Appendix A: Supported Models Reference
Complete list of models available in Azimuth Orion (as of plugin version 1.0):
Note: Not all models may be available for your API key. Access depends on the restrictions and usage tier associated with your key (e.g., free vs. paid, rate limits, model eligibility). If a model fails with a rate limit or access error, try another model or check your provider's dashboard for your key's capabilities.
xAI (Grok)
| Model ID | Display Name | Context Window | Max Output | Vision | Cost Tier | Default Timeout |
|---|---|---|---|---|---|---|
| grok-4 | Grok 4 | 256k tokens | 131k | Yes | High | 90s |
| grok-4-1-fast-reasoning | Grok 4.1 Fast Reasoning | 2M tokens | 131k | Yes | Very Low | 45s |
| grok-4-1-fast-non-reasoning | Grok 4.1 Fast | 2M tokens | 131k | Yes | Very Low | 30s |
| grok-code-fast-1 | Grok Code Fast | 256k tokens | 32k | No | Low | 60s |
Notes:
- grok-code-fast-1: Code-optimized, agentic coding (default for xAI)
- grok-4: Flagship, highest quality reasoning
- grok-4-1-fast-reasoning / grok-4-1-fast-non-reasoning: 2M context, vision support
OpenAI
| Model ID | Display Name | Context Window | Max Output | Vision | Cost Tier | Default Timeout |
|---|---|---|---|---|---|---|
| gpt-5.2 | GPT-5.2 | 400k tokens | 128k | Yes | High | 90s |
| gpt-5-mini | GPT-5 mini | 400k tokens | 128k | Yes | Low | 60s |
| gpt-5-nano | GPT-5 nano | 400k tokens | 128k | Yes | Very Low | 45s |
| gpt-5.3-codex | GPT-5.3 Codex | 400k tokens | 128k | Yes | High | 90s |
| o4-mini | o4-mini | 200k tokens | 100k | Yes | Low | 60s |
Notes:
- gpt-5.2: Best for coding and agentic tasks (flagship)
- gpt-5-mini: Faster, cost-efficient for well-defined tasks
- o4-mini: Fast reasoning model with vision
Anthropic (Claude)
| Model ID | Display Name | Context Window | Max Output | Vision | Cost Tier | Default Timeout |
|---|---|---|---|---|---|---|
| claude-opus-4-6 | Claude Opus 4.6 | 200k tokens | 128k | Yes | Very High | 120s |
| claude-sonnet-4-6 | Claude Sonnet 4.6 | 200k tokens | 64k | Yes | Medium | 75s |
| claude-haiku-4-5 | Claude Haiku 4.5 | 200k tokens | 64k | Yes | Low | 45s |
Notes:
- claude-sonnet-4-6: Best speed/intelligence balance (default for Anthropic, Feb 2026)
- claude-opus-4-6: Most intelligent, Adaptive Thinking
- claude-haiku-4-5: Fastest Claude model, lowest cost
Google (Gemini)
| Model ID | Display Name | Context Window | Max Output | Vision | Cost Tier | Default Timeout |
|---|---|---|---|---|---|---|
| gemini-3-pro-preview | Gemini 3 Pro (Preview) | 1M tokens | 65k | Yes | High | 90s |
| gemini-3-flash-preview | Gemini 3 Flash (Preview) | 1M tokens | 65k | Yes | Low | 60s |
| gemini-3.1-pro-preview | Gemini 3.1 Pro (Preview) | 1M tokens | 65k | Yes | High | 90s |
| gemini-2.5-pro | Gemini 2.5 Pro | 1M tokens | 65k | Yes | Medium | 90s |
| gemini-2.5-flash | Gemini 2.5 Flash | 1M tokens | 65k | Yes | Low | 60s |
Notes:
- gemini-2.5-flash: Price-performance (default for Google)
- gemini-2.5-pro: Advanced reasoning (Stable)
- gemini-3.x: Preview models with 1M context
Legend
Context Window: Maximum input tokens (your question + attached context)
Max Output: Maximum completion tokens (AI response length)
Vision: Supports image input (screenshots)
Cost Tier: Relative pricing (Very Low < Low < Medium < High < Very High)
Default Timeout: Recommended HTTP timeout for this model