#38
Llama 3.3 70B Instruct
llama-3.3-70b-instruct
0.0
Helpfulness
Instruction Following
Comprehension
Empathy
Creative Writing
Helpfulness
0.0
Instruction Following
0.0
Comprehension
0.0
Empathy
0.0
Creative Writing
0.0
Speed
Avg 88 tok/s
Release Date
December 6, 2024
Lab
Meta
Type
Open Source
Context Size
128K
Max Output Tokens
8.2K
Cost per 1 million tokens
$0.23 / $0.40
Model Inputs*
Text
Model Outputs*
Text
Tool Calling*
Enabled
Overall Assistant Score
An average score combining the 5 main categories.
76.93 pts
Rank #38
23th Percentile
0
Novice
33
Capable
66
Proficient
100
Expert
Llama 3.3 70B is a highly optimized version of Meta's open weights model, delivering performance comparable to the larger Llama 3.1 405B model but at a fraction of the cost and size. It excels in complex reasoning, coding, and instruction following, making it one of the most powerful and efficient open models available for general-purpose applications.

Intelligence

Overall Score; Higher is better

Llama 4 Mav…
Gemini 2.5 …
Kimi K2 Thi…
Grok 3 Mini
DeepSeek V3…
Llama 3.3 7…
GPT 4o
GLM 4.6
Llama 4 Sco…
Gemini 2.0 …

Speed

Output Tokens per Second; Higher is better

Gemini 2.5 …
DeepSeek V3…
GLM 4.7 Fla…
Gemini 2.0 …
Grok 4 Fast…
Llama 3.3 7…
Gemini 3.1 …
DeepSeek V3
ChatGPT 4o
GPT 4o

Price

USD per 1M Tokens; Lower is better

Grok 4.1 Fa…
Grok 4.1 Fa…
Grok 4 Fast…
Grok 4 Fast
Llama 4 Mav…
Llama 3.3 7…
GPT-5 Mini
DeepSeek V3
Gemini 2.5 …
Grok 3 Mini

meta Models

Overall Score; Same provider comparison

Llama 4 Mav…
Llama 3.3 7…
Llama 4 Sco…
Llama 3.3 8…

Llama Family

Overall Score; Same model family comparison

Llama 4 Mav…
Llama 3.3 7…
Llama 4 Sco…
Llama 3.3 8…

Closest Rivals

Overall Score; Nearest by overall rank

Llama 4 Mav…
Gemini 2.5 …
Kimi K2 Thi…
Grok 3 Mini
DeepSeek V3…
Llama 3.3 7…
GPT 4o
GLM 4.6
Llama 4 Sco…
Gemini 2.0 …