Model Comparison

Comprehensive side-by-side analysis of model capabilities and performance

Llama 3.1 405B Instruct

Meta

Llama 3.1 405B Instruct is a language model developed by Meta. It achieves strong performance with an average score of 79.2% across 18 benchmarks. It excels particularly in ARC-C (96.9%), GSM8k (96.8%), API-Bank (92.0%). It supports a 256K token context window for handling large documents. The model is available through 8 API providers. Released in 2024, it represents Meta's latest advancement in AI technology.

Llama 3.3 70B Instruct

Meta

Llama 3.3 70B Instruct is a language model developed by Meta. It achieves strong performance with an average score of 79.9% across 9 benchmarks. It excels particularly in IFEval (92.1%), MGSM (91.1%), HumanEval (88.4%). The model shows particular specialization in code tasks with an average performance of 89.4%. It supports a 256K token context window for handling large documents. The model is available through 9 API providers. Released in 2024, it represents Meta's latest advancement in AI technology.

Llama 3.1 405B Instruct

Meta

2024-07-23

Llama 3.3 70B Instruct

Meta

2024-12-06

4 months newer

Pricing Comparison

Cost per million tokens (USD)

Llama 3.1 405B Instruct

Input:$0.89

Output:$0.89

Llama 3.3 70B Instruct

$1.38 cheaper

Input:$0.20

Output:$0.20

Performance Metrics

Context window and performance specifications

Llama 3.1 405B Instruct

Max Context:256.0K

Parameters:405.0B

Llama 3.3 70B Instruct

Max Context:256.0K

Parameters:70.0B

Average performance across 7 common benchmarks

Llama 3.1 405B Instruct

+0.1%

Average Score:78.8%

Llama 3.3 70B Instruct

Average Score:78.6%

Performance comparison across key benchmark categories

Llama 3.1 405B Instruct

code

81.4%

math

+3.4%

87.4%

general

+2.6%

73.2%

Llama 3.3 70B Instruct

code

+7.9%

89.4%

math

84.0%

general

70.7%

Benchmark Scores - Detailed View

Side-by-side comparison of all benchmark scores

Provider Availability & Performance

Available providers and their performance metrics

Llama 3.1 405B Instruct

8 providers

Together

Throughput: 35 tok/s

Latency: 0.5ms

Hyperbolic

Throughput: 40 tok/s

Latency: 0.5ms

DeepInfra

Throughput: 27 tok/s

Latency: 0.5ms

Google

Throughput: 42 tok/s

Latency: 0.4ms

Fireworks

Throughput: 78 tok/s

Latency: 0.5ms

Bedrock

Throughput: 100 tok/s

Latency: 0.5ms

Lambda

Throughput: 42 tok/s

Latency: 0.5ms

Replicate

Throughput: 22 tok/s

Latency: 0.5ms

Llama 3.3 70B Instruct

9 providers

Sambanova

Throughput: 1096 tok/s

Latency: 0.65ms

Together

Throughput: 65 tok/s

Latency: 0.65ms

Hyperbolic

Throughput: 42 tok/s

Latency: 0.65ms

DeepInfra

Throughput: 37 tok/s

Latency: 0.65ms

Fireworks

Throughput: 197 tok/s

Latency: 0.65ms

Groq

Throughput: 268 tok/s

Latency: 0.65ms

Bedrock

Throughput: 100 tok/s

Latency: 0.5ms

Lambda

Throughput: 42 tok/s

Latency: 0.65ms

Cerebras

Throughput: 2220 tok/s

Latency: 0.65ms

Llama 3.1 405B Instruct

+0.1%

Avg Score:78.8%

Providers:8

Llama 3.3 70B Instruct

Avg Score:78.6%

Providers:9