TA/deepseek-ai/DeepSeek-R1-Distill-Llama-70B

Common Name: DeepSeek R1 Distill Llama 70B

Released on Feb 17 12:00 AM

DeepSeek R1 reasoning model distilled to Llama 70B architecture, hosted on TogetherAI.

Context

128K

Inputtext

Outputtext

Collecting…

Input$2.20/MTokens

Output$2.20/MTokens

Availability Trend (24h)

$3.30/$7.70/M

ctx64Kmax8Kavail—tps—

InOut

DeepSeek's reasoning model trained via large-scale reinforcement learning, hosted on TogetherAI.

$1.38/$1.38/M

ctx64Kmax8Kavail—tps—

InOut

DeepSeek V3 MoE model with 671B total parameters and 37B active, hosted on TogetherAI.

$3.85/$3.85/M

ctx128Kmax—avail—tps—

InOut

Meta's largest Llama 3.1 405B model optimized for fast inference on TogetherAI.

$1.32/$1.32/M

ctx128Kmax—avail—tps—

InOut

Alibaba's Qwen2.5 7B model optimized for fast inference on TogetherAI.