Model Insights

mixtral-8x7b-instruct-v0.1

Details

Developer

mistral

License

Apache 2.0

Model parameters

8x7b

Supported context length

32k

Price for prompt token

$0.6/Million tokens

Price for response token

$0.6/Million tokens

Model Performance Across Task-Types

Chainpoll Score

Short Context

0.83

Medium Context

0.99

Model Insights Across Task-Types

Digging deeper, here’s a look how mixtral-8x7b-instruct-v0.1 performed across specific datasets

Short Context RAG

Medium Context RAG

This heatmap indicates the model's success in recalling information at different locations in the context. Green signifies success, while red indicates failure.

Prompt Type

Performance Summary

Tasks	Task insight	Cost insight	Dataset	Context adherence	Avg response length
Short context RAG	The model demonstrates below average reasoning and comprehension skills at short context RAG. It shows just passing mathematical proficiency, as evidenced by its performance on DROP and ConvFinQA benchmarks.	We wish it performed well but its a costly model for performance it offers. We recomend using 6x cheaper Gemini Flash or Haiku.	Drop	0.80	393
			Hotpot	0.81	393
			MS Marco	0.89	393
			ConvFinQA	0.80	393
Medium context RAG	Flawless performance making it suitable for any context length upto 20000 tokens. It struggles a bit for 25000 but can be used without much issues.	Great performance but we recommed using 6x cheaper Gemini Flash.	Medium context RAG	0.99	393