Tokenando
BETA
Subscribe
← Back to prices·All DeepInfra models
DeepInfra
DeepInfraFrontier

WizardLM-2 8x22B (DI)

Fine-tuned

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 (DI) is a frontier AI model from DeepInfra. It costs $0.630 per million input tokens and $0.630 per million output tokens (blended $0.630/M), with a 64,000-token context window.

Profile inherited from upstream Mixtral 8x22B (WizardLM-2 base) ↗ — this is a hosted variant of the same open-weights model.

Released Apr 2024Modalities textOfficial model page ↗Provider pricing ↗API docs ↗Compare with another model →Estimate monthly cost →
INPUT
$0.630/M
per million input tokens
OUTPUT
$0.630/M
per million output tokens
BLENDED 70/30
$0.630/M
unchanged since 3 May
default reference rate · how it is calculated →
CONTEXT
64,000
tokens
What it is good at
  • Open-weights MoE
  • Cheap inference per active parameter
  • 64K context
Typical use cases
  • Self-hosted chat at scale
  • Fine-tune base

Benchmarks

vs. best public score
Scores inherited from Mixtral 8x22B (WizardLM-2 base) — this is a hosted variant of the same open-weights model, so the underlying benchmark scores are identical.
MMLU78%
Multitask academic knowledge across 57 subjects.
GPQA Diamond36%
Graduate-level science questions, "Google-proof".
MATH42%
High-school competition math problems.
HumanEval76%
Python function synthesis from docstrings.
LMArena Elo1198 Elo
Crowd-sourced head-to-head preference Elo rating.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources, treat as directional rather than authoritative.

More from DeepInfra

See all 11 →
Llama 3.3 70B (DI)
Serverless · 128,000 ctx
in $0.230/Mout $0.400/M
Llama 3.1 405B (DI)
Serverless · 128,000 ctx
in $0.800/Mout $0.800/M
Llama 3.1 70B (DI)
Balanced · 128,000 ctx
in $0.350/Mout $0.400/M
Mistral 7B (DI)
Nano · 32,000 ctx
in $0.070/Mout $0.070/M
Mixtral 8x7B (DI)
MoE · 32,000 ctx
in $0.240/Mout $0.240/M
DeepSeek-R1 (DI)
Reasoning · 64,000 ctx
in $0.550/Mout $2.190/M

Frequently asked questions

How much does WizardLM-2 8x22B (DI) cost?

WizardLM-2 8x22B (DI) costs $0.630 per million input tokens and $0.630 per million output tokens, for a blended reference rate of $0.630 per million tokens.

What is WizardLM-2 8x22B (DI)'s context window?

WizardLM-2 8x22B (DI) supports up to 64,000 tokens of context in a single request.

What is WizardLM-2 8x22B (DI) best for?

WizardLM-2 8x22B (DI) is well suited to Open-weights MoE, Cheap inference per active parameter and 64K context.

Who makes WizardLM-2 8x22B (DI)?

WizardLM-2 8x22B (DI) is developed and served by DeepInfra. It was released in Apr 2024.

Terms used on this page

Full glossary →
TokenInput tokensOutput tokensPrice per million tokensContext windowPrompt cachingBatch APIBlended cost