About
DeepSeek R1 Distill Llama 70B is a distilled large language model based on [Llama-3.3-70B-Instruct](/meta-llama/llama-3.3-70b-instruct), using outputs from [DeepSeek R1](/deepseek/deepseek-r1). The model combines advanced distillation techniques to achieve high performance across...
Pricing
Input
$0.80
per 1M tokens
Output
$0.80
per 1M tokens
Blended
$0.80
per 1M tokens
Cheaper than 43% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.80
Monthly
$24.00
vs. Similar Models
Claude 3.5 Sonnet (Oct '24)Q:0.0
$6.00+650%
Qwen3 VL 30B A3B InstructQ:+0.1
$0.23-72%
Hermes 4 - Llama-3.1 70B (Reasoning)Q:+0.1
$0.20-75%
Meta: Llama 4 ScoutQ:+0.1
$0.15-81%
Performance
62
tokens/sec
Faster than 31% of models
0.43
seconds
Faster than 93% of models
32.55
seconds
Faster than 15% of models
Market Median
94 tok/s
34% slower
Median TTFT
1.10s
61% faster
Throughput/Dollar
78
tok/s per $/1M
Speed Comparison
GPT-5.5 (Non-reasoning)
62 tok/s+0%
Qwen3 14B (Reasoning)
62 tok/s-0%
GLM-5.1 (Non-reasoning)
62 tok/s-1%
Context Window
128K
tokens
Larger than 16% of models
Max Output
8K
tokens
6% of context
Benchmarks
MMLU-Pro
79.5%
GPQA Diamond
40.2%
HLE
6.1%
LiveCodeBench
26.6%
SciCode
31.3%
TerminalBench Hard
1.5%
MATH-500
93.5%
AIME
67.0%
AIME 2025
53.7%
IFBench
27.6%
Long Context Recall
11.0%
Tau2
21.9%
Market AverageTop Score