LLM EXPLORER 56,604 MODELS INDEXED

Agents A1 4B Fable Preview by hotdogs

By hotdogs · 103 downloads
HF Repo ↗ Copy URL

Agents A1 4B Fable Preview is an open-source language model by hotdogs. Features: 4b LLM, VRAM: 8.4GB, Context: 256K, License: agpl-3.0, LLM Explorer Score: 0.42, ELO: 1432.

  Agents   Agents-a1 Base model:finetune:internscie... Base model:internscience/agent...   Chatml   Conversational Dataset:hotdogs/uka-fable-reas...   En   Endpoints compatible   Fable   Function-calling   Preview   Qwen   Qwen3 5 text   Reasoning   Region:us   Safetensors   Sft   Th   Tool-call   Tool-use
Model Card on HF 🤗: https://huggingface.co/hotdogs/Agents-A1-4B-Fable-Preview 

Agents A1 4B Fable Preview Benchmarks

LMSys ELO 1432 vs 1272 (so35) 12.6%
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

Agents A1 4B Fable Preview Parameters and Internals

LLM NameAgents A1 4B Fable Preview
Repository 🤗https://huggingface.co/hotdogs/Agents-A1-4B-Fable-Preview 
Base Model(s)  InternScience/Agents-A1-4B   InternScience/Agents-A1-4B
Model Size4b
Required VRAM8.4 GB
Updated2026-07-27
Maintainerhotdogs
Model Typeqwen3_5_text
Model Files  8.4 GB
Supported Languagesen th
Model ArchitectureQwen3_5ForCausalLM
Licenseagpl-3.0
Context Length262144
Model Max Length262144
Transformers Version5.5.0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size248320
Errorsreplace

Best Alternatives to Agents A1 4B Fable Preview

Best Alternatives
Context / RAM
Downloads
Likes
...nts A1 4B Kimi Preview Heretic256K / 8.4 GB2870
Zira Researcher256K / 8.4 GB1017212
Qwen3.5 4B EU256K / 9.7 GB870
Qwen35 4B Soyuz Merged256K / 8.4 GB693
Retrico Lm 4B Finetuned V4256K / 0 GB1430
Holo 3.1 4B Coder256K / 16.8 GB411
Qwen3.5 4B INT8256K / 4.8 GB1112
Distil Qwen35 4B256K / 8.4 GB2222
Qwen3.5 4B INT8256K / 4.8 GB1912
Qwen3.5 4B Abliterated256K / 8.4 GB19911