WizardLM-2 8x22B

Text input Text output
Author's Description

WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model. It demonstrates highly competitive performance compared to leading proprietary models, and it consistently outperforms all existing state-of-the-art opensource models. It is an instruct finetune of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b). To read more about the model release, [click here](https://wizardlm.github.io/WizardLM2/). #moe

Key Specifications
Cost
$$$
Context
65K
Parameters
22B
Released
Apr 15, 2024
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

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

WizardLM-2 8x22B, Microsoft AI's advanced Wizard model, demonstrates highly competitive performance. It consistently ranks among the fastest models and offers among the most competitive pricing, making it an economically attractive option. The model also exhibits exceptional reliability, with a 97th percentile ranking, indicating minimal technical failures and consistent evaluable responses. Across benchmarks, WizardLM-2 shows a mixed performance profile. It excels in Ethics and General Knowledge, achieving 98.0% and 95.5% accuracy respectively, placing it in the 43rd and 48th percentiles for these categories. Email Classification is another strong suit with 95.0% accuracy. However, the model struggles significantly with Instruction Following, scoring 0.0% accuracy, and shows notable weaknesses in Coding (62.0% accuracy, 30th percentile) and Reasoning (43.1% accuracy, 29th percentile). While its speed and cost efficiency are standout strengths, the model's performance in complex logical and programming tasks, as well as its complete failure in instruction following, represent key areas for improvement.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.48
Completion $0.48

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Parasail
Parasail | microsoft/wizardlm-2-8x22b 65K $0.48 / 1M tokens $0.48 / 1M tokens
DeepInfra
DeepInfra | microsoft/wizardlm-2-8x22b 65K $0.48 / 1M tokens $0.48 / 1M tokens
Novita
Novita | microsoft/wizardlm-2-8x22b 65K $0.62 / 1M tokens $0.62 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Free Executions Accuracy Cost Duration
Other Models by microsoft