WizardLM-2 8x22B
Largest open-weights Mixtral MoE. Cheap-to-serve frontier-ish quality before Llama 3.1 405B and DeepSeek V3 took the open-weights lead.
WizardLM-2 8x22B is a efficient AI model from Microsoft. It costs $0.620 per million input tokens and $0.620 per million output tokens (blended $0.620/M), with a 65,535-token context window.
Profile inherited from upstream Mixtral 8x22B (WizardLM-2 base) ↗ — this is a hosted variant of the same open-weights model.
- Open-weights MoE
- Cheap inference per active parameter
- 64K context
- Self-hosted chat at scale
- Fine-tune base
Benchmarks
More from Microsoft
See all 8 →Frequently asked questions
How much does WizardLM-2 8x22B cost?
WizardLM-2 8x22B costs $0.620 per million input tokens and $0.620 per million output tokens, for a blended reference rate of $0.620 per million tokens.
What is WizardLM-2 8x22B's context window?
WizardLM-2 8x22B supports up to 65,535 tokens of context in a single request.
What is WizardLM-2 8x22B best for?
WizardLM-2 8x22B is well suited to Open-weights MoE, Cheap inference per active parameter and 64K context.
Who makes WizardLM-2 8x22B?
WizardLM-2 8x22B is developed and served by Microsoft. It was released in Apr 2024.