LlamaFactory
Product status: Status not recorded
No lifecycle review is recorded for this listing. Its inclusion in the directory does not confirm current availability; check the vendor before relying on it.
No status check date recorded
Apache-2.0 toolkit for configurable language-model fine-tuning, with CLI/WebUI workflows; model weights and training data have separate licenses. No lifecycle review is recorded for this listing. Its inclusion in the directory does not confirm current availability; check the vendor before relying on it.
Product preview
Decision notes
Official-document research
Based on the cited documentation. We did not use the product, purchase a plan, or run quality, speed or reliability tests in this review.
Consider it for
- Developers with training infrastructure and evaluation data
Choose another approach if
- A ready model, free GPU service or guaranteed quality improvement
Pricing and conditions
Documented terms
The Apache-2.0 code license is no-charge. It does not include GPU time, memory/storage, electricity, model/data access or external monitoring services. No universal per-run cost, hosted subscription quote, currency, tax or regional hardware bill was confirmed. Estimate the selected model size, method, sequence length and infrastructure separately.
Pricing source read:
Pricing or terms sourceCode, model and service licensing
Apache-2.0 covers the toolkit with applicable license/NOTICE/change obligations, not every supported weight or dataset. The repository explicitly directs users to each model's own license. Base weights, adapters/merged checkpoints, training data and generated output need separate permission review; no specific model's commercial rights were verified here. Dataset documentation says historical assistant responses can also become SFT training material, so conversation consent matters. Downloads, hub publication and optional telemetry/monitoring providers are separate processing paths, without a project-wide zero-retention or no-training contract established here.
Basis and limits of the recorded score
The existing 4.5 editorial score is retained without its original scoring worksheet. This document review did not re-score or measure performance.
Overview
LlamaFactory configures model training through CLI/YAML and a documented WebUI. SFT documentation includes LoRA, frozen-parameter and full-parameter choices and requires the model, dataset and prompt template to match. Data preparation covers instruction, conversation-history and preference formats. A supported name does not establish that every release, quantization or device configuration works.
This is a training toolkit, not a supplied model, hosted inference subscription or GPU allocation. Select the exact checkpoint and dataset before estimating hardware, running evaluations or publishing an adapter. Optional monitoring, model/data downloads and pushing to a hub introduce separate network paths; local training should not be presented as proof that every configuration stays offline or that training will improve quality.
Sources
The factual claims in this review were read from the pages below. Each date is when an editor last opened the source, not when the product changed.
- Vendor documentation Checked 2026-10-15
- Vendor documentation Checked 2026-10-15
- Vendor documentation Checked 2026-10-15
- Vendor documentation Checked 2026-10-15
- Vendor documentation Checked 2026-10-15
Pros / Cons
Pros
- Explicit training and dataset configuration
- Apache-2.0 toolkit code
Cons
- Compute and compatibility need configuration
- Model/data permissions are not supplied by the toolkit
Who it's for
Domain-data SFT experiments
Adapter training with reviewed datasets
Bottom line
Choose it for a team prepared to configure and evaluate training; verify the exact model and dataset licenses before use or distribution.
Key Features
Tags
Compare LlamaFactory with alternatives
Frequently asked questions
Is LlamaFactory currently available?
No lifecycle review is recorded for this listing. Its inclusion in the directory does not confirm current availability; check the vendor before relying on it.
What is LlamaFactory?
LlamaFactory is a Agents & Developer Tooling tool developed by LlamaFactory. Apache-2.0 toolkit for configurable language-model fine-tuning, with CLI/WebUI workflows; model weights and training data have separate licenses.
What costs and terms should I check for LlamaFactory?
The Apache-2.0 code license is no-charge. It does not include GPU time, memory/storage, electricity, model/data access or external monitoring services. No universal per-run cost, hosted subscription quote, currency, tax or regional hardware bill was confirmed. Estimate the selected model size, method, sequence length and infrastructure separately.
Is LlamaFactory open source?
Apache-2.0 covers the toolkit with applicable license/NOTICE/change obligations, not every supported weight or dataset. The repository explicitly directs users to each model's own license. Base weights, adapters/merged checkpoints, training data and generated output need separate permission review; no specific model's commercial rights were verified here. Dataset documentation says historical assistant responses can also become SFT training material, so conversation consent matters. Downloads, hub publication and optional telemetry/monitoring providers are separate processing paths, without a project-wide zero-retention or no-training contract established here.
What are the best alternatives to LlamaFactory?
Notable Agents & Developer Tooling alternatives include Firecrawl, Browser Use, Daytona. Browse all 33 tools in our Agents & Developer Tooling category for a full overview.
Was this listing useful?
Basic Info
- Pricing
- Free
The Apache-2.0 code license is no-charge. It does not include GPU time, memory/storage, electricity, model/data access or external monitoring services. No universal per-run cost, hosted subscription quote, currency, tax or regional hardware bill was confirmed. Estimate the selected model size, method, sequence length and infrastructure separately.
- Editorial score
- ★ 4.5 / 5 How we score
- Platforms
- LibraryCLIWeb
- Last verified
- Not recorded
