ermiaazarkhalili/FastContext-4B-RL_base-SFT-Fable5

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 7, 2026Architecture:Transformer Featherless Exclusive Cold

The FastContext-4B-RL_base-SFT-Fable5 model by ermiaazarkhalili is a 4 billion parameter Qwen3ForCausalLM architecture, fine-tuned using LoRA on the private Fable-5-Complete-2M-Clean dataset. This model is a supervised fine-tune of the Microsoft FastContext-1.0-4B-RL base model, optimized for instruction-following tasks. It leverages a 4096 maximum sequence length and 4-bit QLoRA precision for efficient deployment.

Loading preview...

Model Overview

ermiaazarkhalili/FastContext-4B-RL_base-SFT-Fable5 is a 4 billion parameter language model based on the Qwen3ForCausalLM architecture. It is a LoRA (Low-Rank Adaptation) fine-tune of the microsoft/FastContext-1.0-4B-RL base model, specifically supervised fine-tuned on the private ermiaazarkhalili/Fable-5-Complete-2M-Clean dataset.

Key Characteristics

  • Architecture: Qwen3ForCausalLM with 4.0 billion parameters.
  • Fine-tuning Method: LoRA using Unsloth and TRL.
  • Training Data: Fine-tuned on a single, private instruction-following dataset (ermiaazarkhalili/Fable-5-Complete-2M-Clean).
  • Efficiency: Utilizes 4-bit QLoRA for base precision and a maximum sequence length of 4096 tokens.
  • Training Details: Trained for 2 epochs with a learning rate of 0.0002 and an effective batch size of 8.

Limitations

  • No Benchmarks: No downstream benchmark evaluations have been conducted; only training loss observations are available.
  • Inherited Biases: Inherits biases, knowledge cutoff, and failure modes from its base model.
  • Untested Behavior: Behavior outside the specific instruction-following distribution it was fine-tuned on is untested.
  • Merged Adapters: LoRA adapters are merged into the base weights, preventing detachment from this fine-tune.