Instructions to use AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5")
model = AutoModelForCausalLM.from_pretrained("AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5

SGLang

How to use AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5 with Docker Model Runner:
```
docker model run hf.co/AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Parable-Granite-4.1-3B-Claude-Fable-5

Granite 4.1 3B trained on real Claude Fable 5 and GPT-5.5 agent traces: 87% lower held-out test loss than its base. A compact chat model trained on the prose side of real agent sessions: strongest at explanations, idioms, and one-liners.

Parable-Granite-4.1-3B is an ibm-granite/granite-4.1-3b fine-tune trained on real multi-step agent sessions: planning, tool use, and <think> reasoning captured from actual Claude Fable 5 and GPT-5.5 agent work, not synthetic Q&A. Smallest and newest release in the Parable series, alongside Parable-Qwen3-4B.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto", device_map="auto")

messages = [{"role": "user", "content": "Write a bash one-liner to find the 10 largest files in a directory tree."}]
inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
output = model.generate(inputs, max_new_tokens=3000, temperature=0.7, top_p=0.95, do_sample=True)
print(tokenizer.decode(output[0][inputs.shape[1]:], skip_special_tokens=True))

Output opens with a <think>...</think> reasoning block before the final answer. Strip it before showing responses to end users.

Sampling: temperature 0.7, top_p 0.95. Budget max_new_tokens generously (at least 2500): trace-trained reasoning models think at length before answering.

GGUF quants for llama.cpp, Ollama, and LM Studio: Parable-Granite-4.1-3B-Claude-Fable-5-GGUF.

Training data

Glint-Research/Fable-5-traces: 4.4k real Claude Fable 5 coding-agent session traces with <think> reasoning and tool calls (AGPL-3.0)
Roman1111111/gpt5.5-terminal: terminal-agent task solutions (MIT)

Every example passed a quality gate (schema validation, secrets scrub, length filtering) before training. QLoRA fine-tune (NF4, sequence length 2048) trained on a single 16 GB GPU, quantized with llama.cpp.

Evaluation

Held-out test split, identical evaluation code and context length for base and fine-tune:

Metric	Base Granite-4.1-3B	Parable	Δ
Test loss	2.824	0.376	−87%

Qualitative review (34 coding/terminal/debugging prompts, strictly graded by mentally executing every answer): 14 of 34 fully correct, 26 of 34 correct or partially correct. The pattern is consistent: reliable on explanations, one-liners, idiomatic refactors, and debugging advice; unreliable on multi-part script and config generation, where we recommend the 8B instead. We publish these numbers because strict qualitative grading is rare in this niche; judge accordingly.

For reference, the strongest published fine-tune on this data family (a 9B) reports 0.71 validation loss. Cross-repo numbers are indicative only: splits, tokenizers, and context lengths differ (ours is measured at 1,024 tokens).

Limitations

Best in its lane: explanations, one-liners, and idiomatic fixes. For multi-step script or config generation, use the 8B; this model can hallucinate agent-transcript formatting on those prompts (3 of 34 in our eval).
Fine-tuned at 2,048-token sequences; the base model's native 128K-token context remains fully available, so long sessions work, with the fine-tuned behavior strongest in the opening turns.

As a fine-tune it inherits Granite-4.1-3B's base behaviors and knowledge cutoff. As with any local model, treat generated commands and code as drafts to review.

Provenance & licensing

Model weights: Apache-2.0 (inherited from Granite-4.1-3B). Training data licenses: Fable-5-traces AGPL-3.0, gpt5.5-terminal MIT. Because those traces originate from third-party assistants, the providers' terms may apply to downstream training and distillation. If you plan to build on this model commercially, confirm your use aligns with those terms.

Get Parable

Platform	Command / Link
Ollama	`ollama run parable/fable` (parable namespace)
Hugging Face	GGUF quants, full weights, eval reports
LM Studio	search "parable" in-app, or any HF GGUF repo URL