NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/WizardLM-2 8x22B
Microsoft logo

WizardLM-2 8x22B

by Microsoft · Released Apr 2024

Open Source
61.4
avg score
Rank #78
Compare
Better than 75% of all models
Context
66K tokens (~33 books)
Input $/1M
$0.62
Output $/1M
$0.62
Type
text
License
Open Source
Benchmarks
7 tested
Data as of
About

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...

Tested on 7 benchmarks · BenchGecko score 61.4. Top scores: IFEval (52.7%), BBH (HuggingFace) (48.6%), Aider — Code Editing (44.4%).

Looking for similar performance at lower cost?
gpt-oss-20b (free) scores 61.0 (99% as good) at $0.00/1M input · 100% cheaper
Capabilities
coding
44.4
#120 globally
reasoning
14.5
#152 globally
math
25.0
#211 globally
knowledge
28.8
#244 globally
general
48.6
#37 globally
language
52.7
#123 globally
Benchmark Scores
Compare All
Tested on 7 benchmarks · Ranked across 6 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
codingCompare coding →
Aider — Code Editing

Code editing benchmark from the Aider project. Measures ability to apply targeted code changes while maintaining correctness and style.

44.4·
reasoningCompare reasoning →
MUSR

HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.

14.5·
mathCompare math →
MATH Level 5

HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.

25.0·
Quick compare:Google DeepMind logovs Gemini 3.5 FlashMicrosoft logovs phi-3-mini 3.8Bvs PaLM 2-S
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Similar Models
Google DeepMind logo
Gemini 3.5 Flash
Google DeepMind
61.3$1.50/1M
Microsoft logo
phi-3-mini 3.8B
Microsoft
61.5n/a
PaLM 2-S
Unknown
61.7n/a
Links
Info
MicrosoftPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Microsoft
Source Code
GitHubHugging Face
BenchGecko API
wizardlm-2-8x22b
Specifications
  • Typetext
  • Context66K tokens (~33 books)
  • ReleasedApr 2024
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.002
Available On
Microsoft logoMicrosoft$0.62
Categories
codingreasoningmathknowledgegenerallanguage
Learn More
transformeropen weightstokens
Share & Export
Tweet
Related Models
Google DeepMind logoGemini 3.5 Flash
61.3
Microsoft logophi-3-mini 3.8B
61.5
PaLM 2-S
61.7
xAI logoGrok 4.3
61.7
WizardLM-2 8x22B is an open-source text AI model by Microsoft, released in April 2024. It has an average benchmark score of 61.4. Context window: 66K tokens.

Related Models

Gemini 3.5 Flash · Google DeepMindphi-3-mini 3.8B · MicrosoftPaLM 2-S · UnknownGrok 4.3 · xAIgpt-oss-20b (free) · OpenAI

Benchmarks

IFEvalBBH (HuggingFace)Aider — Code EditingMMLU-PROMATH Level 5

Related Pages

Microsoft · ProviderMicrosoft · EconomyAll ModelsCompare ModelsPricingDevelopers · API

Key facts · as of 2026-10-05

  • WizardLM-2 8x22B by Microsoft. BenchGecko score 61.4, rank 78 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.62 input · $0.62 output per 1M tokens (as of 2026-10-05).
  • Sold by 1 provider (as of 2026-10-05): Novita (bf16) $0.62 in / $0.62 out.

How to cite · data as of 2026-10-05

WizardLM-2 8x22B · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/wizardlm-2-8x22b

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP