LLM EXPLORER 67,952 MODELS INDEXED

S1 Mini by superwhisper

By superwhisper · 1136 downloads

S1 Mini is an open-source language model by superwhisper. Features: 0.6b LLM, VRAM: 1.5GB, Context: 40K, License: other, LLM Explorer Score: 0.45, ELO: 1337.

  Arxiv:2505.09388   Asr   Automatic-speech-recognition Base model:finetune:qwen/qwen3...   Base model:qwen/qwen3-0.6b   Conversational   Dictation   En   Endpoints compatible   Inverse-text-normalization   Post-processing   Punctuation   Qwen3   Region:us   Safetensors   Speech-to-text   Text-normalization   Truecasing
Model Card on HF πŸ€—: https://huggingface.co/superwhisper/s1-mini 

S1 Mini Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

S1 Mini Parameters and Internals

LLM NameS1 Mini
Repository πŸ€—https://huggingface.co/superwhisper/s1-mini 
Base Model(s)  Qwen/Qwen3-0.6B   Qwen/Qwen3-0.6B
Model Size0.6b
Required VRAM1.5 GB
Updated2026-08-21
Maintainersuperwhisper
Model Typeqwen3
Model Files  1.5 GB
Supported Languagesen
Model ArchitectureQwen3ForCausalLM
Licenseother
Context Length40960
Model Max Length40960
Transformers Version4.51.0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Torch Data Typebfloat16
Errorsreplace

Quantized Models of the S1 Mini

Model
Likes
Downloads
VRAM
...Mini FreeFlow Cleanup MLX 6bit0400 GB

Best Alternatives to S1 Mini

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3 Reranker 0.6B40K / 1.2 GB2508313377
Qwen3 0.6B40K / 1.5 GB13645971
MonkeyGrab Qwen3 0.6B RAG40K / 1.2 GB1120
Wolof Qwen3 0.6B V261040K / 1.2 GB691
StructuraAI40K / 2.4 GB582
Qwen3 0.6B Emoji Gif40K / 1.2 GB3801
RGOO Qwen3 0.6B40K /  GB3341
RGOO Qwen3 0.6B40K /  GB2812
Creo Classify 0.6B40K / 1.2 GB3271
Qwen3 0.6B Tools40K / 1.2 GB26791