NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Qwen2.5 72B Instruct Abliterated

Qwen2.5 72B Instruct Abliterated

by HuiHui AI · Released Oct 2024

87.2
avg score
Rank #10
Compare
Top 3% of all models
Context
N/A
Input $/1M
n/a
Output $/1M
n/a
Type
text-generation
License
Proprietary
Benchmarks
6 tested
Data as of
About

Huihui-ai text generation model. 367K downloads on HuggingFace.

Tested on 6 benchmarks · BenchGecko score 87.2. Top scores: IFEval (85.9%), BBH (HuggingFace) (60.5%), MATH Level 5 (60.1%).

Capabilities
reasoning
12.3
#162 globally
math
60.1
#86 globally
knowledge
34.9
#228 globally
language
85.9
#33 globally
general
60.5
#12 globally
Benchmark Scores
Compare All
Tested on 6 benchmarks · Ranked across 5 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
reasoningCompare reasoning →
MUSR

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

12.3·
mathCompare math →
MATH Level 5

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

60.1·
knowledgeCompare knowledge →
MMLU-PRO

HuggingFace MMLU-Pro. Harder version of MMLU with 10 answer choices instead of 4 and more challenging questions.

50.4·
GPQA

HuggingFace evaluation of GPQA (Graduate-Level Google-Proof Q&A). PhD-level science questions that cannot be easily searched.

19.4·
Quick compare:Anthropic logovs Claude Opus 5Anthropic logovs Claude Fable 5.1DeepSeek logovs DeepSeek-V2 (MoE-236B, May 2024)
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Similar Models
Anthropic logo
Claude Opus 5
Anthropic
86.5$5.00/1M
Anthropic logo
Claude Fable 5.1
Anthropic
88.2$10.00/1M
DeepSeek logo
DeepSeek-V2 (MoE-236B, May 2024)
DeepSeek
85.9n/a
Links
Info
HuiHui AIPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@HuiHui AI
BenchGecko API
huihui-ai-qwen25-72b-instruct-abliterated
Specifications
  • Typetext-generation
  • ContextN/A
  • ReleasedOct 2024
  • LicenseProprietary
  • StatusActive
Available On
HuiHui AIn/a
Categories
reasoningmathknowledgelanguagegeneral
Learn More
transformertokens
Share & Export
Tweet
Related Models
Anthropic logoClaude Opus 5
86.5
Anthropic logoClaude Fable 5.1
88.2
DeepSeek logoDeepSeek-V2 (MoE-236B, May 2024)
85.9
OpenAI logoGPT-5.1-Codex-Max
85.6
Qwen2.5 72B Instruct Abliterated is a proprietary text-generation AI model by HuiHui AI, released in October 2024. It has an average benchmark score of 87.2.

Related Models

Claude Opus 5 · AnthropicClaude Fable 5.1 · AnthropicDeepSeek-V2 (MoE-236B, May 2024) · DeepSeekGPT-5.1-Codex-Max · OpenAIGPT-5 Chat · OpenAI

Benchmarks

IFEvalBBH (HuggingFace)MATH Level 5MMLU-PROGPQA

Related Pages

HuiHui AI · ProviderHuiHui AI · EconomyAll ModelsCompare ModelsPricingDevelopers · API

Key facts · as of 2026-04-09

  • Qwen2.5 72B Instruct Abliterated by HuiHui AI. BenchGecko score 87.2, rank 10 of 312 scored models (normalized average of public benchmark scores).
  • List price n/a input · n/a output per 1M tokens (as of 2026-04-09).

How to cite · data as of 2026-04-09

Qwen2.5 72B Instruct Abliterated · benchmarks, pricing and providers. BenchGecko, data as of 2026-04-09. https://benchgecko.ai/model/huihui-ai-qwen25-72b-instruct-abliterated

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