Prepaid inference-hour packs for open-weight models on Hugging Face, on one checkout: small, mid-size, large, MoE, embedding and legacy families. Each pack is a virtual product - no shipping, and an endpoint key is issued once the order completes.
New
Brand: Hugging Face
DeepSeek V3.2 · 10 hours on H100 - a prepaid inference-hour pack for deepseek-ai/DeepSeek-V3.2, billed at Hugging Face's published Inference Endpoints rate of $4.50 per hour on H100 80GB.What you get10 hour(s) of dedicated H100 80GB inference.The model loaded from the Hub: deepseek-ai/DeepSeek-V3.2...
$40.50
New
Brand: Hugging Face
Dolphin 2.9.1 Yi 34B · 10 hours on H100 - a prepaid inference-hour pack for dphn/dolphin-2.9.1-yi-1.5-34b, billed at Hugging Face's published Inference Endpoints rate of $4.50 per hour on H100 80GB.What you get10 hour(s) of dedicated H100 80GB inference.The model loaded from the Hub: dphn/dolphin-2...
$40.50
New
Brand: Hugging Face
GPT-2 · 10 hours on T4 - a prepaid inference-hour pack for openai-community/gpt2, billed at Hugging Face's published Inference Endpoints rate of $0.50 per hour on T4 16GB.What you get10 hour(s) of dedicated T4 16GB inference.The model loaded from the Hub: openai-community/gpt2.No shipping: the endpo..
$5.00
New
Brand: Hugging Face
gpt-oss 20B · 1 hour on A100 - a prepaid inference-hour pack for openai/gpt-oss-20b, billed at Hugging Face's published Inference Endpoints rate of $2.50 per hour on A100 80GB.What you get1 hour(s) of dedicated A100 80GB inference.The model loaded from the Hub: openai/gpt-oss-20b.No shipping: the en..
$2.50
New
Brand: Hugging Face
gpt-oss 20B · 10 hours on A100 - a prepaid inference-hour pack for openai/gpt-oss-20b, billed at Hugging Face's published Inference Endpoints rate of $2.50 per hour on A100 80GB.What you get10 hour(s) of dedicated A100 80GB inference.The model loaded from the Hub: openai/gpt-oss-20b.No shipping: the..
$25.00
New
Brand: Hugging Face
Llama 3.1 8B Instruct · 1 hour on L4 - a prepaid inference-hour pack for meta-llama/Llama-3.1-8B-Instruct, billed at Hugging Face's published Inference Endpoints rate of $0.80 per hour on L4 24GB.What you get1 hour(s) of dedicated L4 24GB inference.The model loaded from the Hub: meta-llama/Llama-3.1..
$0.80
New
Brand: Hugging Face
Llama 3.1 8B Instruct · 10 hours on L4 - a prepaid inference-hour pack for meta-llama/Llama-3.1-8B-Instruct, billed at Hugging Face's published Inference Endpoints rate of $0.80 per hour on L4 24GB.What you get10 hour(s) of dedicated L4 24GB inference.The model loaded from the Hub: meta-llama/Llama-..
$8.00
New
Brand: Hugging Face
Ornith 1.5 9B · 10 hours on L4 - a prepaid inference-hour pack for ornith-ai/Ornith-1.5-9B-GGUF, billed at Hugging Face's published Inference Endpoints rate of $0.80 per hour on L4 24GB.What you get10 hour(s) of dedicated L4 24GB inference.The model loaded from the Hub: ornith-ai/Ornith-1.5-9B-GGUF...
$8.00
New
Brand: Hugging Face
Qwen2.5 0.5B · 10 hours on T4 - a prepaid inference-hour pack for Qwen/Qwen2.5-0.5B-Instruct, billed at Hugging Face's published Inference Endpoints rate of $0.50 per hour on T4 16GB.What you get10 hour(s) of dedicated T4 16GB inference.The model loaded from the Hub: Qwen/Qwen2.5-0.5B-Instruct.No sh..
$5.00
New
Brand: Hugging Face
Qwen2.5 1.5B · 10 hours on T4 - a prepaid inference-hour pack for Qwen/Qwen2.5-1.5B-Instruct, billed at Hugging Face's published Inference Endpoints rate of $0.50 per hour on T4 16GB.What you get10 hour(s) of dedicated T4 16GB inference.The model loaded from the Hub: Qwen/Qwen2.5-1.5B-Instruct.No sh..
$5.00
New
Brand: Hugging Face
Qwen2.5 7B Instruct · 10 hours on L4 - a prepaid inference-hour pack for Qwen/Qwen2.5-7B-Instruct, billed at Hugging Face's published Inference Endpoints rate of $0.80 per hour on L4 24GB.What you get10 hour(s) of dedicated L4 24GB inference.The model loaded from the Hub: Qwen/Qwen2.5-7B-Instruct.No..
$8.00
New
Brand: Hugging Face
Qwen3 0.6B · 1 hour on T4 - a prepaid inference-hour pack for Qwen/Qwen3-0.6B, billed at Hugging Face's published Inference Endpoints rate of $0.50 per hour on T4 16GB.What you get1 hour(s) of dedicated T4 16GB inference.The model loaded from the Hub: Qwen/Qwen3-0.6B.No shipping: the endpoint key is..
$0.50