Providers

Hamr supports OpenAI-compatible and Anthropic Messages providers. You can configure multiple providers and switch between them at runtime.

Provider Protocols

Protocol	Description
openai-compatible	Any service with an OpenAI-compatible /chat/completions endpoint
anthropic-compatible	Real Anthropic Messages API adapter via /v1/messages

Built-in Provider Presets

Hamr ships with presets for these providers:

Provider ID	Protocol	Base URL	Auth
relay	openai-compatible	http://127.0.0.1:1234/v1	none
custom	openai-compatible	user-configured	optional
deepseek	openai-compatible	https://api.deepseek.com/v1	DEEPSEEK_API_KEY
openrouter	openai-compatible	https://openrouter.ai/api/v1	OPENROUTER_API_KEY
groq	openai-compatible	https://api.groq.com/openai/v1	GROQ_API_KEY
anthropic	anthropic-compatible	https://api.anthropic.com	ANTHROPIC_API_KEY
mistral	openai-compatible	https://api.mistral.ai/v1	MISTRAL_API_KEY
together	openai-compatible	https://api.together.xyz/v1	TOGETHER_API_KEY

Configuring a Provider

All examples use the current multi-provider config format. The legacy [provider] format from v0.1–v0.3 still works but is not recommended for new configs.

Local Relay

[active]
provider = "relay"
model = "Qwen3.6-35B-A3B-UD-IQ3_XXS.gguf"

[providers.relay]
enabled = true
base_url = "http://127.0.0.1:1234/v1"

The base_url line is only needed when not running on the default port. Set model to the exact ID your server reports from GET /models.

DeepSeek

[active]
provider = "deepseek"
model = "deepseek-chat"

[providers.deepseek]
enabled = true
api_key_env = "DEEPSEEK_API_KEY"

OpenRouter

[active]
provider = "openrouter"
model = "deepseek/deepseek-chat"

[providers.openrouter]
enabled = true
api_key_env = "OPENROUTER_API_KEY"

[providers.openrouter.headers]
HTTP-Referer = "https://github.com/skaft-software/hamr"
X-Title = "Hamr"

Groq

[active]
provider = "groq"
model = "llama-3.3-70b-versatile"

[providers.groq]
enabled = true
api_key_env = "GROQ_API_KEY"

Anthropic

[active]
provider = "anthropic"
model = "claude-sonnet-4-5-20250929"

[providers.anthropic]
enabled = true
compatibility = "anthropic-compatible"
api_key_env = "ANTHROPIC_API_KEY"

Anthropic uses the real Messages API (POST /v1/messages) with x-api-key auth. System prompts map to the top-level system field. Tool use uses the native Anthropic tool format.

Custom OpenAI-compatible

[active]
provider = "myserver"
model = "local-model"

[providers.myserver]
enabled = true
base_url = "http://127.0.0.1:8080/v1"

[[providers.myserver.models]]
id = "local-model"
display_name = "My Local Model"
context_window = 131072
supports_thinking = false

API Key Configuration

Prefer api_key_env over api_key. The environment variable is never written to disk.

[providers.deepseek]
api_key_env = "DEEPSEEK_API_KEY"  # reads from process.env.DEEPSEEK_API_KEY

Custom Headers

Header values can reference environment variables using $VAR or ${VAR} syntax:

[providers.relay.headers]
"CF-Access-Client-Id" = "${CF_ACCESS_CLIENT_ID}"
"CF-Access-Client-Secret" = "${CF_ACCESS_CLIENT_SECRET}"

Hamr resolves variables from process.env at runtime. Unset vars omit the header entirely.

Token Pricing & Session Spend

Cloud providers have preset token pricing. Override in config:

[providers.deepseek]
input_price_per_1m_tokens = 0.27   # USD per 1M input tokens
output_price_per_1m_tokens = 1.10  # USD per 1M output tokens

Default pricing per provider:

Provider	Input ($/1M)	Output ($/1M)
DeepSeek	$0.27	$1.10
Groq	$0.59	$0.79
Anthropic	$3.00	$15.00
OpenRouter	varies	varies

When pricing is configured, the TUI shows a session spend indicator at the bottom: Spend: $0.004. Local providers show Spend: local.

Switching Providers

From the TUI: press /, type settings, navigate to the Model tab. Changes persist to the active config file.

Live Provider Smoke Tests

Smoke tests verify end-to-end connectivity. Run from source only — they require API keys and are not run in CI.

HAMR_LIVE_PROVIDER=relay bun run smoke:provider
HAMR_LIVE_PROVIDER=deepseek bun run smoke:provider
HAMR_LIVE_PROVIDER=openrouter bun run smoke:provider
HAMR_LIVE_PROVIDER=anthropic bun run smoke:provider
HAMR_LIVE_PROVIDER=custom HAMR_CUSTOM_BASE_URL=http://127.0.0.1:1234/v1 bun run smoke:provider

Known Limitations

• Mistral and Together are available as presets but lack dedicated smoke tests.
• Streaming is implemented in the shared OpenAI-compatible client but the Anthropic adapter only supports non-streaming requests.