Arxiv:2305.18290 Autotrain compatible Base model:finetune:pirxtion/m... Base model:pirxtion/mnlp m2 dp... Conversational Dpo Endpoints compatible Generated from trainer Qwen3 Region:us Safetensors Trl Unsloth
Model Card on HF 🤗: https://huggingface.co/PirxTion/tulu-mathqa
Tulu Mathqa 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)
Tulu Mathqa Parameters and Internals
| LLM Name | Tulu Mathqa |
| Repository 🤗 | https://huggingface.co/PirxTion/tulu-mathqa |
| Model Name | tulu-mathqa |
| Base Model(s) | |
| Model Size | 596m |
| Required VRAM | 1.2 GB |
| Updated | 2025-06-09 |
| Maintainer | PirxTion |
| Model Type | qwen3 |
| Model Files | |
| Model Architecture | Qwen3ForCausalLM |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.52.3 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|vision_pad|> |
| Vocabulary Size | 151936 |
| Torch Data Type | bfloat16 |
| Errors | replace |
Best Alternatives to Tulu Mathqa
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Sam Reason A1 | 40K / 2.4 GB | 39 | 1 |
| SFT Nochat FULL DATA | 32K / 1.2 GB | 424 | 0 |
| Fft Qwen | 32K / 0 GB | 129 | 0 |
| Sft Scp Epoch1 | 32K / 1.2 GB | 405 | 0 |
| MNLP M3 Mcqa Model | 32K / 1.2 GB | 84 | 0 |
| Qwen3 Wiki Sciq Mmlu | 32K / 1.2 GB | 73 | 0 |
| MNLP M3 Mcqa Model | 32K / 1.2 GB | 27 | 1 |
| ... 15000 B4 2E 512T LR1e 05 ACC4 | 32K / 2.4 GB | 134 | 0 |
| ...DPO Model Smoltalk Bigger Test | 32K / 1.2 GB | 39 | 0 |
| Full Dataset Instruction V4 | 32K / 1.2 GB | 17 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than PirxTion/tulu-mathqa .
Rank the Tulu Mathqa 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? 48046 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124