Home / Pricing

One live rate for every model.

No plans or seats. Pay per token at a platform-set per-model rate, the same no matter which Mac serves the request.

pay per token · from $0.03/M input · 12 models live

Up to 10M free tokens

$1 promotional credit on signup. No card required.

The real token count depends on the model and your input/output mix, so treat 10 million as an estimate. See the complete calculation.

Start free

Why idle Macs change the cost structure

Typical API pricing includes several layers between silicon and your app: capacity is bought, rented, repackaged, and metered. Umbra routes demand to Apple Silicon that is already paid for. The marginal cost is mostly electricity, which is how the network can undercut commodity hosted rates on the long-tail catalog without inventing subsidies.

Providers earn a majority share of metered on-demand revenue at the same platform-set rates you pay. Cash-out eligibility is in the provider agreement.

Live per-model rates

Priced per million tokens. Output is typically higher than input.

GET /v1/models

Model	Quant	Input $/M	Output $/M	Min RAM
hf-deepreinforce-ai-ornith-1.0-35b-gguf-ornith-1-0-35b-q4-k-m	Q4_K_M	$0.40/M	$1.20/M	23 GB
hf-deepreinforce-ai-ornith-1.0-9b-gguf-ornith-1-0-9b-q4-k-m	Q4_K_M	$0.10/M	$0.30/M	8 GB
hf-empero-ai-qwythos-9b-claude-mythos-5-1m-gguf-qwythos-9b-claude-mythos-5-1m-mtp-q4-k-m	Q4_K_M	$0.10/M	$0.30/M	16 GB
hf-hauhaucs-qwen3.6-35b-a3b-uncensored-hauhaucs-aggressive-qwen3-6-35b-a3b-uncensored-hauhaucs-aggressive-q4-k-m	Q4_K_M	$0.40/M	$1.20/M	23 GB
hf-liquidai-lfm2.5-230m-gguf-lfm2-5-230m-bf16	Q4_0_4_4	$0.03/M	$0.10/M	8 GB
hf-liquidai-lfm2.5-230m-gguf-lfm2-5-230m-q4-k-m	Q4_K_M	$0.03/M	$0.10/M	8 GB
hf-liquidai-lfm2.5-230m-gguf-lfm2-5-230m-q8-0	Q8_0	$0.03/M	$0.10/M	8 GB
hf-qwen-qwen2.5-0.5b-instruct-gguf-qwen2-5-0-5b-instruct-fp16	F16	$0.03/M	$0.10/M	8 GB
hf-qwen-qwen3-8b-gguf-qwen3-8b-q4-k-m	Q4_K_M	$0.10/M	$0.30/M	8 GB
hf-thebloke-tinyllama-1.1b-chat-v1.0-gguf-tinyllama-1-1b-chat-v1-0-q2-k	Q2_K	$0.10/M	$0.30/M	8 GB
hf-unsloth-gemma-3-4b-it-gguf-gemma-3-4b-it-q4-k-m	Q4_K_M	$0.10/M	$0.30/M	8 GB
hf-yuxinlu1-gemma-4-12b-coder-fable5-composer2.5-v1-gguf-gemma4-coding-q4-k-m	Q4_K_M	$0.20/M	$0.60/M	9 GB

Alpha honesty: prices may change during the alpha. Pricing is platform-set and identical for every provider serving a model; providers do not set their own rates. How free credit works.

For developers

Keep your OpenAI or Anthropic SDK, change the base URL and key, pay only for tokens used. Free credit on signup, no card required. Read the quickstart.

Create an account

For providers

Earn per token from these same platform-set prices, a majority share of metered on-demand revenue. Cash out via the provider wallet when available; eligibility is in the provider agreement.

See host economics