You can deploy this model and start earning money today!

hotdogs/
Agents-A1-4B-Fable-Preview

Name: Agents-A1-4B-Fable-Preview
Availability: InStock

Views

No views yet

🤖 Agents-A1-4B-Fable-Preview

4B Agent Model (Text-only) — SFT Fine-tuned · Fable Reasoning · Tool-Calling

Preview release — Fine-tuned from InternScience/Agents-A1-4B (Qwen3.5-based VLM, text decoder only) on high-quality Fable-style reasoning traces. Features agentic reasoning with think blocks, tool-use patterns, and multi-turn conversation support.

Evaluation

SWE-bench Verified (subset)

Metric	Value
Resolve rate	49.5% (99/200)
Instances evaluated	200 / 500 (first 200, index-ordered slice)
Scaffold	mini-swe-agent v2.4.6
Agent config	step_limit=250, cost_limit=$3.0, temperature=0.0
Inference	llama.cpp, F16, ctx=131072
Empty patches	13/200 (6.5%)
Date	2026-08-02

Note: This is evaluated on a 200-instance subset (first 200 by dataset index, not a stratified random sample), not the full 500-instance SWE-bench Verified set. Results may differ from a full-set evaluation. Full results and prediction files available at [link if you publish them].

✨ Key Features

Capability	Description
🧠 Fable Reasoning	Step-by-step CoT with `<think>` blocks
🔧 Tool Calling	Hermes/Qwen function-calling format
💬 Multi-turn	Trained on full agent trajectories
🌏 Thai + English	Native bilingual support
💻 Code & Shell	Python, bash, system tasks
⚡ Fast Inference	4B params — runs on consumer GPUs (6-8 GB VRAM)

🚀 Usage

Python (Transformers)

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "hotdogs/Agents-A1-4B-Fable-Preview",
    torch_dtype="auto",
    device_map="auto",
    trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("hotdogs/Agents-A1-4B-Fable-Preview")

messages = [{"role": "user", "content": "Hello"}]
inputs = tokenizer.apply_chat_template(messages, tokenize=True, return_tensors="pt")
outputs = model.generate(inputs, max_new_tokens=256, temperature=0.6)
print(tokenizer.decode(outputs[0]))

GGUF (llama.cpp)

For GGUF quantized version with vision support (via mmproj), see: 👉 hotdogs/Agents-A1-4B-Fable-Preview-GGUF

📦 Downloads

File	Size	Quant	Description
`Agents-A1-4B-Fable-Preview (Safetensors)`	8.41 GB	BF16	Full precision (Transformers)
GGUF versions	2.6–8.4 GB	IQ4_NL to f16	See GGUF repo

🧬 Architecture

Parameter	Value
Base	InternScience/Agents-A1-4B
Parameters	~4.29B
Hidden Size	2,560
Attention	Linear + Standard hybrid (8 full-attn layers)
Context	8,192 tokens (trained on 4,096)
Precision	BF16 / GGUF quantized
Format	ChatML (Jinja2 template)
Vocab	248,320 tokens

Note: This is the text decoder only (Qwen3_5ForCausalLM). For vision capabilities, use the GGUF version with mmproj from the base model.

🏋️ Training Details

Base Model

InternScience/Agents-A1-4B — A 4B VLM based on Qwen3.5 architecture with hybrid linear/full attention. We fine-tuned the text decoder only.

Dataset

hotdogs/uka-fable-reasoning/perfect-v1 — 3,500 high-quality multi-turn agent reasoning traces, featuring:

Fable-5 style think blocks
Tool-use trajectories
Multi-turn conversations
System-prompt guided interactions
Quality-controlled (QC passed)

Training Pipeline

┌─────────────────────────────────────────────────────────────┐
│                    Training Pipeline                         │
├─────────────────────────────────────────────────────────────┤
│                                                             │
│  Base Model                     LoRA Adapter                │
│  (Agents-A1-4B VLM)             (r=64, α=128)              │
│       │                              │                     │
│       └───────────┬──────────────────┘                     │
│                   │                                        │
│         Unsloth FastLanguageModel                           │
│         (text decoder only)                                │
│         device_map="auto" · 4× RTX 3060 12GB               │
│                   │                                        │
│                   ▼                                        │
│    ┌──────────────────────────┐                            │
│    │  Custom SFTDataset        │                            │
│    │  • apply_chat_template()  │                            │
│    │  • Tokenize (4096 max)    │                            │
│    │  • Mask prompt tokens     │                            │
│    └──────────┬───────────────┘                            │
│               │                                            │
│               ▼                                            │
│    ┌──────────────────────────┐                            │
│    │  HF Trainer (no SFTTr.)   │                            │
│    │  • BF16 · LR=1e-4        │                            │
│    │  • max_grad_norm=1.0     │                            │
│    │  • 3 epochs · batch=1    │                            │
│    │  • ~9 hours on 4×3060    │                            │
│    └──────────┬───────────────┘                            │
│               │                                            │
│               ▼                                            │
│         Final Model (scale=0.35)                           │
│    (LoRA merged into text decoder)                         │
│                                                             │
└─────────────────────────────────────────────────────────────┘

Hardware

Component	Spec
GPUs	4× NVIDIA RTX 3060 12GB
Total VRAM	48 GB
Training Time	~9 hours 20 min
Framework	Unsloth + HuggingFace Trainer

Hyperparameters

Parameter	Value
LoRA Rank	64
LoRA Alpha	128
Learning Rate	1e-4
Scheduler	Cosine (warmup 3%)
Optimizer	AdamW 8-bit
Max Grad Norm	1.0
Sequence Length	4,096
Batch Size	1 per GPU (eff. 4)
Epochs	3
Gradient Checkpointing	✅
Precision	BF16

✅ What This Model Excels At

Agent tasks — Tool calling, planning, multi-step reasoning
Reasoning — Think-block style chain-of-thought
Coding — Python, shell scripts, system tasks
Multi-turn — Full conversation trajectories
Thai + English — Native bilingual capability

⚠️ Limitations

4B model — Smaller than 27B/72B models, may not match their depth
Preview — Early release, more training data planned
Text-only — This release is the text decoder only. For vision, use the GGUF version with mmproj.

💖 Support / โปรดสนับสนุน

If you find this model useful, please consider supporting my work!
หากคุณคิดว่าโมเดลนี้มีประโยชน์ กรุณาสนับสนุนผลงานของฉันด้วยนะคะ! 🙏

₿ Bitcoin — BTC:

bc1qf27cyk3vmugcdyv9xdtuv5jwz37863crpj5c9v

Thank you for your support! 🙏✨
ขอบคุณมากๆ สำหรับการสนับสนุนค่า! 💖🤗

🙏 Acknowledgements / ขอบคุณ

InternScience — For the Agents-A1-4B base model
Qwen Team (Alibaba) — For the Qwen3.5 architecture
Unsloth AI — For the training optimizations
All dataset contributors and the open-source AI community ❤️

Built with ❤️ by UKA — 18-year-old coder & cybersecurity expert