Lambda
This host has discontinued its inference service. Its offerings are excluded from current prices and comparisons; price history is kept. Inference API wound down. lambda.ai/inference (checked 2026-10-04): "As the Inference API winds down, you can continue deploying and scaling models seamlessly on NVIDIA GPU instances." No end date is published, so none is recorded. LiteLLM still lists lambda_ai/* prices; they are not live.
Price history
Blended ($/Mtok, 75% input / 25% output) price over time for every model this host serves. Click a legend entry to hide or show its line.
Hosted offerings
| Model | Variant | Input $/Mtok | Output $/Mtok | Confidence |
|---|---|---|---|---|
| eu.meta.llama3-2-3b-instruct-v1:0 | – | $0.0150 | $0.0250 | medium |
| Llama 3.1 8B Instruct | – | $0.0250 | $0.0400 | medium |
| Llama 4 Maverick | – | $0.0500 | $0.10 | medium |
| Llama 4 Scout | – | $0.0500 | $0.10 | medium |
| Qwen3 32B | – | $0.0500 | $0.10 | medium |
| R1 0528 | – | $0.20 | $0.60 | medium |
| DeepSeek V3 0324 | – | $0.20 | $0.60 | medium |