Llama is a first-party provider, serving models it develops itself. Its published rate is the reference other channels are measured against, and first-party hosting usually comes with the most predictable availability.
It lists 7 models on this page. 7 are offered at $0. Coverage spans text and chat (7). The lowest-priced model hosted here is Llama-3.3-70B-Instruct.
| Cheapest here? | |||||
|---|---|---|---|---|---|
| Llama-3.3-70B-Instructllama-3.3-70b-instruct | Free | — | 128,000 | — | |
| Llama 4 Maverick 17B 128E Instruct FP8llama-4-maverick-17b-128e-instruct-fp8 | Free | — | 128,000 | — | |
| Llama-4-Scout-17B-16E-Instruct-FP8llama-4-scout-17b-16e-instruct-fp8 | Free | — | 128,000 | Only channel | |
| Cerebras-Llama-4-Maverick-17B-128E-Instructcerebras-llama-4-maverick-17b-128e-instruct | Free | — | 128,000 | Only channel | |
| Cerebras-Llama-4-Scout-17B-16E-Instructcerebras-llama-4-scout-17b-16e-instruct | Free | — | 128,000 | Only channel | |
| Groq-Llama-4-Maverick-17B-128E-Instructgroq-llama-4-maverick-17b-128e-instruct | Free | — | 128,000 | Only channel | |
| Llama-3.3-8B-Instructllama-3.3-8b-instruct | Free | — | 128,000 | Only channel |