MiroThinker 1.7 Mini is an open-source language model by miromind-ai. Features: 235b LLM, VRAM: 61.1GB, Context: 256K, License: apache-2.0.
| LLM Name | MiroThinker 1.7 Mini |
| Repository ๐ค | https://huggingface.co/miromind-ai/MiroThinker-1.7-mini |
| Base Model(s) | |
| Model Size | 235b |
| Required VRAM | 61.1 GB |
| Updated | 2026-03-27 |
| Maintainer | miromind-ai |
| Model Type | qwen3_moe |
| Model Files | |
| Supported Languages | en |
| Model Architecture | Qwen3MoeForCausalLM |
| License | apache-2.0 |
| Context Length | 262144 |
| Model Max Length | 262144 |
| Transformers Version | 4.51.1 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151936 |
| Torch Data Type | bfloat16 |
| Errors | replace |
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...en3 235B A22B Instruct 2507 1M | 986K / 171.9 GB | 12 | 1 |
| Qwen3 235B A22B Instruct 2507 | 256K / 171.9 GB | 149338 | 763 |
| Qwen3 235B A22B Thinking 2507 | 256K / 179.9 GB | 65585 | 357 |
| ...n3 235B A22B Instruct 2507 FP8 | 256K / 236.5 GB | 157287 | 124 |
| ...n3 235B A22B Thinking 2507 FP8 | 256K / 236.5 GB | 40776 | 82 |
| ...n3 235B A22B Instruct 2507 FP8 | 256K / 236.5 GB | 4354 | 3 |
| Qwen3 235B A22B Instruct 2507 | 256K / 167.9 GB | 158 | 9 |
| Qwen3 235B A22B Thinking 2507 | 256K / 167.9 GB | 167 | 2 |
| ...n3 235B A22B Thinking 2507 FP8 | 256K / 236.5 GB | 18 | 2 |
| ... 235B A22B Instruct 2507 NVFP4 | 256K / 139.2 GB | 3882 | 6 |
๐ 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! ๐