modelgrep
M

Meta: Llama 3 70B Instruct

meta-llama/llama-3-70b-instruct

169th smartest of 178JSON
Use via OpenRouter ↗
Intelligence
8.9
169th of 178
Design Elo
Speed
18
271st fastest
Latency
1.3s
first token
Input price
$0.510
167th cheapest
Context
8K
8K max out

How it compares

Smarter than5%
of all ranked models
Faster than8%
of all ranked models
Cheaper than44%
of all ranked models

Overview

Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors. This 70B instruct-tuned version was optimized for high quality dialogue usecases. It has demonstrated strong...

Benchmarks

independent · via OpenRouter
Artificial Analysis7th percentile
Intelligence Index
8.9
Coding Index
6.8
Agentic Index
0.0
GPQA Diamond
38%
Humanity's Last Exam
4%
SciCode
19%
Tau²-Bench (agentic)
0%

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
Novitafp8$0.510$0.740100%

Specifications

Context window8K
Max output8K
Knowledge cutoffDec 2023
Input modalitiestext
Output modalitiestext
Prompt caching
Cache read price
ModeratedNo

Llama 3 70B Instruct FAQ

How much does Llama 3 70B Instruct cost?

Llama 3 70B Instruct costs $0.510 per million input tokens and $0.740 per million output tokens via OpenRouter, making it 167th cheapest of 298 paid models.

How smart is Llama 3 70B Instruct?

Llama 3 70B Instruct scores 8.9 on the Artificial Analysis Intelligence Index, ranking 169th of 178 benchmarked models, with a GPQA Diamond score of 38%.

How fast is Llama 3 70B Instruct?

Llama 3 70B Instruct generates around 18 tokens per second with 1.3s time-to-first-token (p50), the 271st fastest tracked model.

What is Llama 3 70B Instruct's context window?

Llama 3 70B Instruct supports a 8K-token context window and can output up to 8K tokens. It accepts text input.

Compare head-to-head