The best Fine-Tuning & Model Training
Every option ranked — open-source, self-hostable, and commercial — by our transparent Sovereignty Score, with honest trade-offs so you choose what fits you, not us.
Adapting a base model to your own data. The managed services charge per training token and keep the resulting weights on their platform; the open toolchains produce adapter files you own outright and can run anywhere.
L4Models & tooling — layer 4 of the AI stack- 1
Axolotl
Top pickOpen sourceFine-tune most open models from one YAML file. The community default.
Free. You rent or own the GPU; a small LoRA can cost a few dollars of rented time. · in our OpenAI Fine-Tuning comparison →
What is Axolotl? →93sovereignty - 2
Unsloth
Open sourceSame fine-tune, far less VRAM. Turns 'we need a bigger GPU' into 'this fits'.
Free and open source; a paid managed tier exists for multi-GPU convenience. · in our OpenAI Fine-Tuning comparison →
What is Unsloth? →92sovereignty - 3
LLaMA-Factory
Open sourceA web UI for fine-tuning. The gentlest on-ramp if YAML is the blocker.
Free and unlimited; hardware is yours. · in our OpenAI Fine-Tuning comparison →
What is LLaMA-Factory? →91sovereignty - 4
torchtune
Open sourcePyTorch-native recipes you can actually read and modify.
Free. Part of the PyTorch ecosystem. · in our OpenAI Fine-Tuning comparison →
What is torchtune? →90sovereignty - 5
Hugging Face PEFT
Open sourceThe adapter library underneath most of the others. Maximum control.
Free, Apache-2.0. · in our OpenAI Fine-Tuning comparison →
What is Hugging Face PEFT? →90sovereignty
Is it free?
The free tier, its real limits and where you start paying — read from each vendor's own pricing page and dated.
Run these yourself
What each one actually needs — real RAM, honest running cost, and the setup time nobody quotes.
- Self-hosting Axolotl 24 GB of VRAM for comfortable LoRA or QLoRA on 7–14B models. Full fine-tuning an 8B is listed at 48 GB on one GPU, and 70B QLoRA+FSDP as "Dual GPU @ 21GB VRAM". Plenty of system RAM too — the FAQ says exit code -9 means you ran out of system RAM, not VRAM.
Replacing a specific tool?
Head-to-head comparisons for each popular fine-tuning & model training product.
Straight head-to-heads
Two fine-tuning & model training tools, side by side — verified facts and a plain verdict.