8 Meta (Llama) models, sorted cheapest-first. Normalized to $/1M tokens from the OpenRouter live feed.
8 Meta (Llama) models ranked by blended cost (3× input + output / 4). Cheapest is Meta: Llama 3.1 8B Instruct at $0.058/M blended. Prices update daily.
Subscribe — $9/mo| Model | Input $/M | Output $/M | Blended $/M |
|---|---|---|---|
| Meta: Llama 3.1 8B Instructcache meta-llama | $0.050/M | $0.080/M | $0.058/M |
| Meta: Llama 3.2 1B Instruct meta-llama | $0.027/M | $0.201/M | $0.070/M |
| Meta: Llama 3.2 3B Instruct meta-llama | $0.051/M | $0.335/M | $0.122/M |
| Meta: Llama 4 Scoutvision meta-llama | $0.100/M | $0.300/M | $0.150/M |
| Meta: Llama 3.3 70B Instruct meta-llama | $0.100/M | $0.320/M | $0.155/M |
| Meta: Llama Guard 4 12Bvision meta-llama | $0.180/M | $0.180/M | $0.180/M |
| Meta: Llama 4 Maverickvisioncache meta-llama | $0.188/M | $0.652/M | $0.304/M |
| Meta: Llama 3.1 70B Instruct meta-llama | $0.400/M | $0.400/M | $0.400/M |
Pro subscribers get emailed the instant a model they track drops in price, plus a weekly cheapest-per-task digest and full CSV/JSON export.
Subscribe — $9/mo