Skip to main content

Model Providers

20+ supported LLM providers with streaming, tool calling, and extended reasoning

k8s-autopilot works with 20+ model providers through LangChain's provider abstraction. You can switch models mid-conversation without restarting, and the agent automatically detects the right provider from the model name and available API keys.


Model Format

Models are referenced using the standard provider:model format:

google_genai:gemini-3.7-flash
anthropic:claude-opus-4-7
openai:gpt-5.4
deepseek:deepseek-reasoner
ollama:llama3.1

You can set the model in several ways:

  • UI Model Picker: Click the model selector in the chat interface to switch instantly.
  • Settings UI: Go to Settings → Auth & Keys to add provider keys, then select models directly in chat.
  • Environment Variable: Set MODEL=gemini-3.7-flash in your .env file.
  • Provider + Model: Set both MODEL and MODEL_PROVIDER for explicit control.
Auto-Detection

If you just set the model name without specifying a provider, k8s-autopilot auto-detects the provider. For example, MODEL=gpt-4o automatically routes to OpenAI, and MODEL=gemini-3.7-flash routes to Google GenAI.


Supported Providers

ProviderIDAuth Env VarHighlights
Google GenAIgoogle_genaiGOOGLE_API_KEYGemini 3.8/3.7/3.6/3.5 Flash, Gemini 3.1 Pro, 1M context, Native Thinking
AnthropicanthropicANTHROPIC_API_KEYClaude 3.7 Sonnet, Claude Opus 4.7/5, Extended Thinking budgets, Vision
OpenAIopenaiOPENAI_API_KEYGPT-5.5, GPT-5.4, o3/o3-mini Reasoning, Vision, Tool calling
Google Vertex AIgoogle_vertexaiADC (GOOGLE_CLOUD_PROJECT)Enterprise Vertex AI endpoints with Workload Identity
Azure OpenAIazure_openaiAZURE_OPENAI_API_KEYSovereign enterprise Azure-hosted OpenAI models
AWS BedrockbedrockAWS IAM Roles / ADCAnthropic Claude and Amazon Nova via AWS private endpoints
GroqgroqGROQ_API_KEYUltra-low latency LPU inference with Llama 3.3 70B
DeepSeekdeepseekDEEPSEEK_API_KEYDeepSeek V3 and DeepSeek R1 reasoning models
Together AItogetherTOGETHER_API_KEYOpen-source foundation models hosted on fast clusters
Fireworks AIfireworksFIREWORKS_API_KEYHigh-speed function calling and quantized open weights
OpenRouteropenrouterOPENROUTER_API_KEYMulti-provider routing gateway (Claude, Gemini, Kimi K3)
Mistral AImistralaiMISTRAL_API_KEYMistral Large, Codestral, Pixtral
NVIDIA NIMnvidiaNVIDIA_API_KEYAccelerated NIM microservice endpoints
PerplexityperplexityPPLX_API_KEYSearch-augmented Sonar models
CoherecohereCOHERE_API_KEYCommand R / Command R+
IBM watsonxibmWATSONX_APIKEYEnterprise Granite and Llama deployments
HuggingFacehuggingfaceHUGGINGFACEHUB_API_TOKENDedicated Inference Endpoints
LiteLLMlitellmLITELLM_API_KEYUnified proxy for internal enterprise gateways
xAIxaiXAI_API_KEYGrok 4.5 and Grok 2 models
BasetenbasetenBASETEN_API_KEYCustom deployed foundation models
OllamaollamaOptional / None100% offline local inference on CPU or GPU

Special Authentication Cases

  • Google Vertex AI and AWS Bedrock use implicit cloud authentication (Application Default Credentials or IAM Role bindings). No explicit API key is needed.
  • Ollama requires no API key at all. Point the agent to the Ollama endpoint and go.

Auto-Detection Rules

When you set MODEL without specifying MODEL_PROVIDER, the agent infers the provider from the model name:

Model Name PrefixDetected Provider
gpt-, o1, o3, o4, chatgptopenai
claude-, sonnet, opus, haikuanthropic
geminigoogle_genai
deepseekdeepseek
llamaollama

If the name doesn't match any prefix, the agent checks which providers have API keys configured, using this priority order:

Google GenAI ──► Anthropic ──► OpenAI ──► Groq ──► DeepSeek ──► OpenRouter

These models have specialized capability and token profiles built into k8s-autopilot:

ModelProviderContext WindowReasoning ControlsTool Calling
Gemini 3.8 Flashgoogle_genai / google_vertexai1M tokens✅ low / medium / high
Gemini 3.7 Flashgoogle_genai / google_vertexai1M tokens✅ low / medium / high
Gemini 3.6 Flashgoogle_genai1M tokens✅ low / medium / high
Gemini 3.5 Flashgoogle_genai1M tokens✅ low / medium / high
Gemini 3.1 Progoogle_genai1M tokens✅ low / high
Gemini 2.5 Progoogle_genai1M tokens✅ low / medium / high
Claude Opus 4.7 / 4.8 / 5anthropic200K tokens✅ low / med / high / max
Claude 3.7 Sonnetanthropic200K tokens✅ low / med / high / max
Claude Sonnet 4.5 / 4.6 / 5anthropic200K tokens✅ low / med / high / max
Claude Haiku 4.5anthropic200K tokens
Claude 3.5 Sonnetanthropic200K tokens
GPT-5.5 / 5.5 Proopenai400K tokens✅ low / medium / high
GPT-5.4 / 5.4 miniopenai400K tokens✅ low / medium / high
o3 / o3-miniopenai200K tokens✅ low / medium / high
GPT-4o / 4o miniopenai128K tokens
DeepSeek V3deepseek64K tokens
DeepSeek R1deepseek64K tokens
GLM 5.2fireworks / baseten / openrouter128K tokens
Grok 4.5 / Grok 2xai131K tokens
Llama 3.3 70Bgroq / together128K tokens
Kimi K3 / K2.7openrouter / baseten1M tokens✅ low / medium / high

Any model from any supported provider works — unlisted models simply use safe framework defaults.


Extended Thinking / Reasoning

k8s-autopilot supports extended reasoning for models with native thinking capabilities. The reasoning effort controls how much thinking the model performs before generating output — higher effort yields more thorough operational diagnostics but consumes more tokens.

Setting the Effort

From .env:

REASONING_EFFORT=high    # Options: low, medium, high, max

From the UI: Go to Settings → Runtime Config to change the reasoning effort.

How It Works Per Provider

The REASONING_EFFORT setting is translated into each provider's native parameters:

ProviderNative ParameterExample for high
Google GenAIthinking_level, thinking_budgetthinking_level="HIGH", thinking_budget=8192
Anthropicthinking.type, thinking.budget_tokensthinking={"type": "enabled", "budget_tokens": 8192}
OpenAIreasoning.effortreasoning={"effort": "high"}

Claude models support max effort while most other providers cap at high. The agent automatically clamps to the nearest supported level.


Custom Base URLs

For self-hosted, proxied, or enterprise deployments, you can override the API endpoint for any provider:

ProviderBase URL Variable(s)
OpenAIOPENAI_BASE_URL, OPENAI_API_BASE
AnthropicANTHROPIC_BASE_URL, ANTHROPIC_API_URL
Google GenAIGOOGLE_GEMINI_BASE_URL
Azure OpenAIAZURE_OPENAI_ENDPOINT
GroqGROQ_BASE_URL, GROQ_API_BASE
DeepSeekDEEPSEEK_API_BASE
FireworksFIREWORKS_BASE_URL, FIREWORKS_API_BASE
TogetherTOGETHER_API_BASE
OpenRouterOPENROUTER_API_BASE
MistralMISTRAL_BASE_URL
NVIDIANVIDIA_BASE_URL
PerplexityPERPLEXITY_BASE_URL
xAIXAI_API_BASE
BasetenBASETEN_BASE_URL, BASETEN_API_BASE
CohereCO_API_URL
HuggingFaceHF_INFERENCE_ENDPOINT
IBM watsonxWATSONX_URL

You can also set custom base URLs per provider directly from the UI by expanding Advanced Settings in the API key modal.


Google Vertex AI

To use Google's enterprise Vertex AI instead of Google AI Studio:

GOOGLE_GENAI_USE_VERTEXAI=true
GOOGLE_CLOUD_PROJECT=your-gcp-project-id
GOOGLE_CLOUD_LOCATION=us-central1

Authentication uses Application Default Credentials (ADC). Ensure you have run gcloud auth application-default login or configured a service account key.


Local Models with Ollama

Run 100% offline and air-gapped with no API keys:

# Pull a model locally
ollama pull llama3.1

# Configure k8s-autopilot to use it
MODEL=llama3.1
MODEL_PROVIDER=ollama

By default, k8s-autopilot connects to Ollama at http://localhost:11434. To target a remote Ollama server, set the OLLAMA_HOST environment variable or configure it in Settings → Auth & Keys → Ollama → Advanced Settings.

Model Size Recommendation

Tool calling and structured outputs work best with models of 13B parameters or larger. Smaller models may struggle with complex multi-step plans.


Next Steps