Model Insights

mixtral-8x22b-instruct-v0.1

Details

Developer

mistral

License

Apache 2.0

Model parameters

176b

Supported context length

64k

Price for prompt token

$1.2/Million tokens

Price for response token

$1.2/Million tokens

Model Performance Across Task-Types

Chainpoll Score

Short Context

0.93

Medium Context

0.99

Model Insights Across Task-Types

Digging deeper, here’s a look how mixtral-8x22b-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 exceptional reasoning and comprehension skills, excelling at short context RAG. It shows good mathematical proficiency, as evidenced by its performance on DROP and ConvFinQA benchmarks.	It is a great open source model but 15x costlier than Gemini Flash with similar performance. We do not recommend using this.	Drop	0.93	368
			Hotpot	0.89	368
			MS Marco	0.94	368
			ConvFinQA	0.96	368
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 12x cheaper Gemini Flash.	Medium context RAG	0.99	368