Arxiv:1910.09700 Autotrain compatible Base model:deepseek-ai/deepsee... Base model:quantized:deepseek-... Conversational Dataset:open-thoughts/openthou... Dataset:openai/gsm8k En Endpoints compatible Gguf Grpo Quantized Qwen2 Region:us Safetensors Sft Trl Unsloth
Model Card on HF 🤗: https://huggingface.co/mrdayl/OpenThink
OpenThink Benchmarks
MMLU Pro: 9.45
GPQA: 4.36
MUSR: 2.01
BBH: 9.18
IFEval: 20.54 vs 88 (so35)-76.7%
MATH Lvl 5: 28.85
LLME Score: 0.2966
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)
OpenThink Parameters and Internals
| LLM Name | OpenThink |
| Repository 🤗 | https://huggingface.co/mrdayl/OpenThink |
| Base Model(s) | |
| Model Size | 1.5b |
| Required VRAM | 1.1 GB |
| Updated | 2025-03-26 |
| Maintainer | mrdayl |
| Model Type | qwen2 |
| Model Files | |
| Supported Languages | en |
| GGUF Quantization | Yes |
| Quantization Type | gguf |
| Model Architecture | AutoModelForCausalLM |
| License | apache-2.0 |
| Model Max Length | 131072 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <|vision_pad|> |
Best Alternatives to OpenThink
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| AceMath 1.5B Instruct 1epoch | 0K / 0.1 GB | 8 | 0 |
| Acemath 200 | 0K / 0.1 GB | 8 | 0 |
| Qwen First Aid Medium | 0K / 1.9 GB | 0 | 0 |
Rank the OpenThink 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? 47753 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227