Skip to main content
Back to Explore

Llama 3.1 Instruct 70B

Meta·Released 2024-07-23
Open SourceMultimodal

Pricing

Input

$0.56

per 1M tokens

Output

$0.56

per 1M tokens

Blended

$0.56

per 1M tokens

Cheaper than 49% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.56

Monthly

$16.80

vs. Similar Models

Ministral 3 3BQ:0.0
$0.10-82%
Qwen3 30B A3B (Non-reasoning)Q:0.0
$0.13-76%
Qwen3 4B (Non-reasoning)Q:0.0
$0.19-66%
IBM: Granite 4.1 8BQ:-0.1
$0.06-89%

Performance

35

tokens/sec

Faster than 4% of models

0.58

seconds

Faster than 79% of models

0.58

seconds

Faster than 85% of models

Market Median

94 tok/s

63% slower

Median TTFT

1.11s

48% faster

Throughput/Dollar

63

tok/s per $/1M

Speed Comparison

Kimi K2.5 (Non-reasoning)
35 tok/s+0%
Claude 4 Opus (Non-reasoning)
35 tok/s+1%
Gemma 4 31B (Reasoning)
35 tok/s+1%

Benchmarks

MMLU-Pro
67.6%
GPQA Diamond
40.9%
HLE
4.6%
LiveCodeBench
23.2%
SciCode
26.7%
TerminalBench Hard
3.0%
MATH-500
64.9%
AIME
17.3%
AIME 2025
4.0%
IFBench
34.4%
Long Context Recall
6.3%
Tau2
15.2%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models