SmallThinker 4BA0.6B Instruct is an open-source language model by PowerInfer. Features: 4b LLM, VRAM: 8.5GB, Context: 32K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.21.
Arxiv:2507.20984 Conversational Custom code En Feature-extraction Instruct Region:us Safetensors Sharded Smallthinker Tensorflow
Model Card on HF 🤗: https://huggingface.co/PowerInfer/SmallThinker-4BA0.6B-Instruct
SmallThinker 4BA0.6B Instruct Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
SmallThinker 4BA0.6B Instruct Parameters and Internals
| LLM Name | SmallThinker 4BA0.6B Instruct |
| Repository 🤗 | https://huggingface.co/PowerInfer/SmallThinker-4BA0.6B-Instruct |
| Model Name | smallthinker_4b_instruct |
| Model Size | 4b |
| Required VRAM | 8.5 GB |
| Updated | 2025-09-23 |
| Maintainer | PowerInfer |
| Model Type | smallthinker |
| Instruction-Based | Yes |
| Model Files | |
| Supported Languages | en |
| Model Architecture | SmallThinkerForCausalLM |
| License | apache-2.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.53.3 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151936 |
| Torch Data Type | bfloat16 |
| Errors | replace |
Best Alternatives to SmallThinker 4BA0.6B Instruct
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| SmallThinker 4BA0.6B Instruct | 32K / 8.5 GB | 290 | 62 |
Note: green Score (e.g. "73.2") means that the model is better than PowerInfer/SmallThinker-4BA0.6B-Instruct .
Rank the SmallThinker 4BA0.6B Instruct 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! 🌟
What open-source LLMs or SLMs are you in search of? 55446 in total.
Original data from HuggingFace, Arena and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a