Benched.ai
View
About
Command Palette
Search for a command to run...
Models
DeepSeek R1 Distill Llama 70B
DeepSeek
Reasoning
Open
Context
Release Date
Jan 20, 2025
Window
128k
Pricing
Per 1M tokens
Input
$0.7
Output
$0.8
Blended 3:1
$0.8
Capabilities
Speed
64 t/s
Input
Output
Reasoning tokens
Latency
TTFT
0.54 ms
500 token response
39.33 s
Benchmarks
Reasoning
●●○○○
Math
●●●●○
Coding
●○○○○
MMLU Pro
79.5%
GPQA
40.2%
HLE
6.1%
SciCode
31.2%
AIME
67.0%
MATH 500
93.5%
LiveCodeBench
26.6%
HumanEval
97.2%
DeepSeek
R1 Distill Llama 70B