Codius App
Connect one organization key in the App, then launch supported coding agents with eligible Codius models selected automatically.
Codius Coding Plans
One coding plan. Every Codius surface.
Pro and Max provide integrated hosted coding-model access for Codius App, supported agents, and the API. Codius App remains free and open source.
Included surfaces
Model access that follows your workflow.
Choose the interface that fits the task without rebuilding provider configuration every time.
Compatible agents
Connect supported coding agents through documented provider configuration instead of maintaining a separate model account for each surface.
OpenAI-compatible API
Use organization-scoped API keys with compatible SDKs, tools, and custom development workflows.
Organization ownership
API keys, subscriptions, member access, and usage belong to an organization with explicit permissions.
Visible usage windows
Track current five-hour, weekly, and billing-cycle usage from the authenticated Codius account.
Extra usage
Eligible organizations can use wallet credit for usage beyond the amount included with a plan.
Commercial model
Free clients. Optional hosted models.
- Codius App
- Open source and free to run with supported agents and providers.
- Supported agents
- Agents launched through Codius App can receive the connected Codius provider and default model without editing their config files.
- Pro and Max
- Paid Coding Plans that include hosted model access and organization usage controls.
- API access
- Organization-scoped keys for compatible tools and custom integrations, subject to the active plan and usage policy.
Catalog
The current Codius model lineup.
Every active Codius model is listed here with its public API ID, limits, and supported capabilities. Plan access can still depend on your organization and usage policy.
Claude Fable 5
anthropic/claude-fable-5
- Context window
- 1M tokens
- Max output
- 128K tokens
Agent compatibleToolsReasoningVision
Claude Haiku 4.5
anthropic/claude-haiku-4.5
- Context window
- 200K tokens
- Max output
- 64K tokens
Agent compatibleToolsReasoningVision
Claude Opus 4.7
anthropic/claude-opus-4.7
- Context window
- 1M tokens
- Max output
- 128K tokens
Agent compatibleToolsReasoningVision
Claude Opus 4.8
anthropic/claude-opus-4.8
- Context window
- 1M tokens
- Max output
- 128K tokens
Agent compatibleToolsReasoningVision
Claude Sonnet 4.6
anthropic/claude-sonnet-4.6
- Context window
- 1M tokens
- Max output
- 128K tokens
Agent compatibleToolsReasoningVision
DeepSeek-V4-Flash-0731
deepseek/v4-flash
- Context window
- 1.0M tokens
- Max output
- 384K tokens
Agent compatibleToolsReasoning
DeepSeek-V4-Pro
deepseek/v4-pro
- Context window
- 1.0M tokens
- Max output
- 384K tokens
Agent compatibleToolsReasoning
Gemini 3.1 Flash Lite
google/gemini-3.1-flash-lite
- Context window
- 1.0M tokens
- Max output
- 66K tokens
Agent compatibleToolsReasoningVision
Gemini 3.1 Pro
google/gemini-3.1-pro
- Context window
- 1.0M tokens
- Max output
- 66K tokens
Agent compatibleToolsReasoningVision
Gemini 3.5 Flash
google/gemini-3.5-flash
- Context window
- 1.0M tokens
- Max output
- 66K tokens
Agent compatibleToolsReasoningVision
Gemini 3.5 Flash-Lite
google/gemini-3.5-flash-lite
- Context window
- 1.0M tokens
- Max output
- 66K tokens
Agent compatibleToolsVision
Gemini 3.6 Flash
google/gemini-3.6-flash
- Context window
- 1.0M tokens
- Max output
- 66K tokens
Agent compatibleToolsReasoningVision
Gemini 3 Flash
google/gemini-3-flash
- Context window
- 1.0M tokens
- Max output
- 66K tokens
Agent compatibleToolsReasoningVision
Gemma 4 31B
google/gemma-4-31b
- Context window
- 262K tokens
- Max output
- 66K tokens
Agent compatibleToolsReasoningVision
MiniMax M2.7
minimax/m2.7
- Context window
- 205K tokens
- Max output
- 131K tokens
Agent compatibleTools
MiniMax M2.7 Highspeed
minimax/m2.7-highspeed
- Context window
- 205K tokens
- Max output
- 131K tokens
Agent compatibleTools
MiniMax M3
minimax/m3
- Context window
- 1M tokens
- Max output
- 512K tokens
Agent compatibleTools
Kimi K2.6
moonshotai/kimi-k2.6
- Context window
- 262K tokens
- Max output
- 131K tokens
Agent compatibleToolsReasoningVision
Kimi K3
moonshotai/kimi-k3
- Context window
- 1.0M tokens
- Max output
- 1.0M tokens
Agent compatibleToolsReasoningVision
GPT-5.4
openai/gpt-5.4
- Context window
- 1.1M tokens
- Max output
- 128K tokens
Agent compatibleToolsReasoningVision
GPT-5.4 Mini
openai/gpt-5.4-mini
- Context window
- 400K tokens
- Max output
- 128K tokens
Agent compatibleToolsReasoningVision
GPT-5.4 Nano
openai/gpt-5.4-nano
- Context window
- 400K tokens
- Max output
- 128K tokens
Agent compatibleToolsReasoningVision
GPT-5.4 Pro
openai/gpt-5.4-pro
- Context window
- 1.1M tokens
- Max output
- 128K tokens
ReasoningVision
GPT-5.5
openai/gpt-5.5
- Context window
- 1.1M tokens
- Max output
- 128K tokens
Agent compatibleToolsReasoningVision
GPT-5.6 Luna
openai/gpt-5.6-luna
- Context window
- 1.1M tokens
- Max output
- 128K tokens
Vision
GPT-5.6 Sol
openai/gpt-5.6-sol
- Context window
- 1.1M tokens
- Max output
- 128K tokens
ReasoningVision
GPT-5.6 Terra
openai/gpt-5.6-terra
- Context window
- 1.1M tokens
- Max output
- 128K tokens
ReasoningVision
GPT-5 Mini
openai/gpt-5-mini
- Context window
- 400K tokens
- Max output
- 128K tokens
Agent compatibleToolsReasoningVision
Grok 4.3
xai/grok-4.3
- Context window
- 1.0M tokens
- Max output
- 131K tokens
Agent compatibleToolsReasoning
GLM-5.1
zai/glm-5.1
- Context window
- 200K tokens
- Max output
- 131K tokens
Agent compatibleToolsReasoning
GLM-5.2
zai/glm-5.2
- Context window
- 1M tokens
- Max output
- 131K tokens
Agent compatibleToolsReasoning
Add model access when you need it.
Compare the live Pro and Max catalog, then continue through the existing authenticated checkout flow.