Autotrain compatible Base model:quantized:qwen/qwen... Base model:qwen/qwen2.5-0.5b-i... Conversational Cot En Endpoints compatible Instruct Onnx Qwen2 Reasoner Region:us Safetensors
Model Card on HF 🤗: https://huggingface.co/prithivMLmods/FastThink-0.5B-Tiny
FastThink 0.5B Tiny Benchmarks
MMLU Pro: 7.21
GPQA: 1.45
MUSR: 3.58
BBH: 5.02
IFEval: 25.8 vs 88 (so35)-70.7%
MATH Lvl 5: 2.04
LLME Score: 0.25857
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)
FastThink 0.5B Tiny Parameters and Internals
| LLM Name | FastThink 0.5B Tiny |
| Repository 🤗 | https://huggingface.co/prithivMLmods/FastThink-0.5B-Tiny |
| Base Model(s) | |
| Model Size | 0.5b |
| Required VRAM | 1 GB |
| Updated | 2025-07-25 |
| Maintainer | prithivMLmods |
| Model Type | qwen2 |
| Instruction-Based | Yes |
| Model Files | |
| Supported Languages | en |
| Model Architecture | Qwen2ForCausalLM |
| License | apache-2.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.47.1 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|PAD_TOKEN|> |
| Vocabulary Size | 151936 |
| Torch Data Type | float16 |
| Errors | replace |
Best Alternatives to FastThink 0.5B Tiny
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Qwen2 0.5B Abyme Merge3 | 128K / 1.3 GB | 2 | 1 |
| Qwen2 0.5B Abyme Merge2 | 128K / 0.3 GB | 6 | 0 |
| Qwen2.5 0.5B Instruct | 64K / 1 GB | 19 | 0 |
| Qwen2.5 0.5B Instruct | 32K / 1 GB | 3012733 | 17 |
| Qwen2.5 0.5B Instruct | 32K / 1 GB | 929542 | 346 |
| Limbic Tool Use 0.5B 32K | 32K / 1 GB | 96 | 17 |
| Qwen2 0.5B Instruct | 32K / 1 GB | 144148 | 190 |
| Qwen2.5 0.5B Instruct | 32K / 1 GB | 156053 | 11 |
| ...5B Instruct With Output Tokens | 32K / 2 GB | 673 | 0 |
| Qwen2.5 Coder 0.5B Instruct | 32K / 1 GB | 12612 | 49 |
Note: green Score (e.g. "73.2") means that the model is better than prithivMLmods/FastThink-0.5B-Tiny .
Rank the FastThink 0.5B Tiny 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? 50068 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124