LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Kimi K2 Thinking by unsloth

 »  All LLMs  »  unsloth  »  Kimi K2 Thinking   URL Share it on

  150   

Kimi K2 Thinking is an open-source language model by unsloth. Features: LLM, VRAM: 383.2GB, Context: 256K, License: other, LLM Explorer Score: 0.23.

Base model:moonshotai/kimi-k2-... Base model:quantized:moonshota...   Compressed-tensors   Conversational   Custom code   Deepseek v3   Endpoints compatible   Region:us   Safetensors   Sharded   Tensorflow   Unsloth
Model Card on HF 🤗: https://huggingface.co/unsloth/Kimi-K2-Thinking 

Kimi K2 Thinking Benchmarks

LLME Score: 0.22903
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Kimi K2 Thinking (unsloth/Kimi-K2-Thinking)
🌟 Advertise your project 🚀

Kimi K2 Thinking Parameters and Internals

LLM NameKimi K2 Thinking
Repository 🤗https://huggingface.co/unsloth/Kimi-K2-Thinking 
Base Model(s)  Kimi K2 Thinking   moonshotai/Kimi-K2-Thinking
Required VRAM383.2 GB
Updated2026-05-08
Maintainerunsloth
Model Typedeepseek_v3
Model Files  1.0 GB: 1-of-62   9.8 GB: 2-of-62   9.8 GB: 3-of-62   9.8 GB: 4-of-62   9.8 GB: 5-of-62   9.8 GB: 6-of-62   9.8 GB: 7-of-62   9.8 GB: 8-of-62   9.8 GB: 9-of-62   9.8 GB: 10-of-62   9.8 GB: 11-of-62   9.8 GB: 12-of-62   9.8 GB: 13-of-62   9.8 GB: 14-of-62   9.8 GB: 15-of-62   9.8 GB: 16-of-62   9.8 GB: 17-of-62   9.8 GB: 18-of-62   9.8 GB: 19-of-62   9.8 GB: 20-of-62   9.8 GB: 21-of-62   9.8 GB: 22-of-62   9.8 GB: 23-of-62   9.8 GB: 24-of-62   9.8 GB: 25-of-62   9.8 GB: 26-of-62   9.8 GB: 27-of-62   9.8 GB: 28-of-62   9.8 GB: 29-of-62   9.8 GB: 30-of-62   9.8 GB: 31-of-62   9.8 GB: 32-of-62   9.8 GB: 33-of-62   9.8 GB: 34-of-62   9.8 GB: 35-of-62   9.8 GB: 36-of-62   9.8 GB: 37-of-62   9.8 GB: 38-of-62   9.8 GB: 39-of-62   9.8 GB: 40-of-62
Model ArchitectureDeepseekV3ForCausalLM
Licenseother
Context Length262144
Model Max Length262144
Transformers Version4.57.3
Tokenizer ClassTikTokenTokenizer
Padding Token[PAD]
Vocabulary Size163840
Torch Data Typebfloat16

Best Alternatives to Kimi K2 Thinking

Best Alternatives
Context / RAM
Downloads
Likes
Kimi K2 Thinking NVFP4256K / 210 GB2724330
Kimi K2 Thinking BF16256K / 220 GB2456
Kimi K2 Instruct 0905 BF16256K / 1399.1 GB1424
Kimi K2 Instruct 0905256K / 667.4 GB487
DeepSeek V3 Base160K / 171.8 GB23451684
DeepSeek V3.1160K / 180.4 GB147941819
DeepSeek V3.1160K / 176.1 GB743
DeepSeek R1 Block INT8160K / 176.1 GB124049
DeepSeek V3160K / 171.8 GB112613
DeepSeek V3.0324160K / 376 GB398
Note: green Score (e.g. "73.2") means that the model is better than unsloth/Kimi-K2-Thinking.

Rank the Kimi K2 Thinking Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 53727 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer