Same open weights, served by 2 providers.
| Provider | Their model id | Serving | Input | Output | Cached in | Blended 3:1 | Source | |
|---|---|---|---|---|---|---|---|---|
DeepInfra | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | flexFP80.8× standard | $0.32 | $0.32 | — | $0.32 | Verified 6h ago · Changed 8h ago | |
DeepInfra Deprecated, 2026-06-11 | meta-llama/Meta-Llama-3.1-70B-Instruct | standardFull precision | $0.40 | $0.40 | — | $0.40 | Verified 6h ago · Changed 8h ago | |
DeepInfra | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | standardFP8 | $0.40 | $0.40 | — | $0.40 | Verified 6h ago · Changed 8h ago | |
AWS Bedrock | meta.llama3-1-70b-instruct-v1:0 | standardQuantization not disclosed | $0.72 | $0.72 | — | $0.72 | Verified 8h ago · Changed 8h ago | |
AWS Bedrock | us.meta.llama3-1-70b-instruct-v1:0 | standardQuantization not disclosed | $0.72 | $0.72 | — | $0.72 | Verified 8h ago · Changed 8h ago |
Same weights don't always mean identical outputs: quantization, context limits and serving stacks can change quality, especially for tool calling. Prices are USD per 1M tokens; hyperscaler prices are for the reference region shown.
| DeepInfra bfloat16 | $48/mo |
| DeepInfra fp8 | $48/mo |
| AWS Bedrock | $86.4/mo |
| AWS Bedrock | $86.4/mo |
Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.