DeepInfra Llama-3.3-70B-Instruct-Turbo model page (list price and Flex tier)

Fields

id

sources/svc-src-deepinfra-llama33-turbo-page-2026-10-17

URL

https://deepinfra.com/meta-llama/Llama-3.3-70B-Instruct-Turbo

Publisher

DeepInfra

Kind

database

Accessed

2026-10-17

notes

Fetched 2026-10-17. Page header: $0.10 in / $0.32 out per 1M tokens. Tier table: Flex (0.8x) $0.08 in / $0.256 out. Prose describes the model as an FP8-quantized faster variant of Llama 3.3 70B. Agrees with the same day's models-list API (0.10/0.32, context 131072) and with the 2026-10-16 Artificial Analysis FAQ split for DeepInfra Turbo FP8.

Notes

Fetched 2026-10-17. Page header: $0.10 in / $0.32 out per 1M tokens. Tier table: Flex (0.8x) $0.08 in / $0.256 out. Prose describes the model as an FP8-quantized faster variant of Llama 3.3 70B. Agrees with the same day's models-list API (0.10/0.32, context 131072) and with the 2026-10-16 Artificial Analysis FAQ split for DeepInfra Turbo FP8.

Cross-references

This record references no other records.