Nous: Hermes 2 Mixtral 8x7B DPO

Text input Text output Unavailable
Author's Description

Nous Hermes 2 Mixtral 8x7B DPO is the new flagship Nous Research model trained over the [Mixtral 8x7B MoE LLM](/models/mistralai/mixtral-8x7b). The model was trained on over 1,000,000 entries of primarily [GPT-4](/models/openai/gpt-4) generated data, as well as other high quality data from open datasets across the AI landscape, achieving state of the art performance on a variety of tasks. #moe

Key Specifications
Cost
$$$$
Context
32K
Parameters
56B
Released
Jan 15, 2024
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Logit Bias Stop Min P Top P Max Tokens Frequency Penalty Temperature Presence Penalty
Performance Summary

Nous Hermes 2 Mixtral 8x7B DPO demonstrates competitive performance in terms of both speed and cost, ranking in the 43rd and 45th percentile respectively across five benchmarks. This indicates it offers a balanced profile for operational efficiency. In terms of accuracy, the model exhibits varied performance across different categories. It achieves its highest accuracy in General Knowledge at 86.0%, placing it in the 26th percentile, suggesting a solid foundation in diverse factual domains. However, its performance in other areas is notably lower. For instance, in Ethics and Email Classification, it scores 53.0% (15th percentile) and 89.0% (13th percentile) respectively, indicating potential areas for improvement in nuanced ethical reasoning and precise categorization. A significant weakness is observed in Coding, where it scores only 7.0% accuracy (13th percentile), suggesting this is not a strong suit for the model. Instruction Following also presents a challenge, with an accuracy of 28.0% (29th percentile). Overall, Nous Hermes 2 Mixtral 8x7B DPO is a cost-effective and reasonably fast model with a strong grasp of general knowledge. Its primary areas for development lie in specialized tasks such as coding, ethical reasoning, and complex instruction following.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.6
Completion $0.6

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Together
Together | nousresearch/nous-hermes-2-mixtral-8x7b-dpo 32K $0.6 / 1M tokens $0.6 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by nousresearch