Arxiv:2305.18290 Autotrain compatible Base model:finetune:qwen/qwen3... Base model:qwen/qwen3-1.7b Conversational Dataset:kevinshin/wildchat-cre... Dpo Endpoints compatible Generated from trainer Qwen3 Region:us Safetensors Sharded Tensorflow Trl
Model Card on HF 🤗: https://huggingface.co/kevinshin/qwen3-1.7b-rpo-lr-1e-5-alpha-1-beta-0.1-wc-cw-3k-neg-rethink-pos
Qwen3 1.7B Rpo Lr 1e 5 Alpha 1 Beta 0.1 Wc Cw 3K Neg Rethink Pos 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)
Qwen3 1.7B Rpo Lr 1e 5 Alpha 1 Beta 0.1 Wc Cw 3K Neg Rethink Pos Parameters and Internals
| LLM Name | Qwen3 1.7B Rpo Lr 1e 5 Alpha 1 Beta 0.1 Wc Cw 3K Neg Rethink Pos |
| Repository 🤗 | https://huggingface.co/kevinshin/qwen3-1.7b-rpo-lr-1e-5-alpha-1-beta-0.1-wc-cw-3k-neg-rethink-pos |
| Model Name | qwen3-1.7b-rpo-lr-1e-5-alpha-1-beta-0.1-wc-cw-3k-neg-rethink-pos |
| Base Model(s) | |
| Model Size | 1.7b |
| Required VRAM | 6.9 GB |
| Updated | 2025-09-17 |
| Maintainer | kevinshin |
| Model Type | qwen3 |
| Model Files | |
| Model Architecture | Qwen3ForCausalLM |
| Context Length | 40960 |
| Model Max Length | 40960 |
| Transformers Version | 4.55.0.dev0 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151936 |
| Torch Data Type | float32 |
| Errors | replace |
Best Alternatives to Qwen3 1.7B Rpo Lr 1e 5 Alpha 1 Beta 0.1 Wc Cw 3K Neg Rethink Pos
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Lucy 128K | 128K / 3.4 GB | 1627 | 106 |
| Polaris 1.7B Preview | 128K / 3.4 GB | 38 | 7 |
| Qwen3 1.7B | 40K / 4 GB | 933199 | 257 |
| Jan V1 Edge | 40K / 3.4 GB | 394 | 32 |
| Smoothie Qwen3 1.7B | 40K / 3.4 GB | 61276 | 1 |
| Lucy | 40K / 3.4 GB | 455 | 63 |
| Qwen3 1.7B | 40K / 3.4 GB | 18651 | 5 |
| Luth 1.7B Instruct | 40K / 3.4 GB | 342 | 10 |
| Qwen3 1.7B | 40K / 3.4 GB | 4644 | 3 |
| Kimina Prover Distill 1.7B | 40K / 4.1 GB | 1629 | 8 |
Note: green Score (e.g. "73.2") means that the model is better than kevinshin/qwen3-1.7b-rpo-lr-1e-5-alpha-1-beta-0.1-wc-cw-3k-neg-rethink-pos .
Rank the Qwen3 1.7B Rpo Lr 1e 5 Alpha 1 Beta 0.1 Wc Cw 3K Neg Rethink Pos 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? 51415 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124