Fine Tuning as a Service (FTAAS)
register → orchestrate → train (HF / TRL / LLaMA-Factory / Unsloth / Axolotl · LoRA / QLoRA / DoRA) → MLflow → vLLM / Ray Serve → API / Console
ftaas.Client / notebook Console
\ /
▼ ▼
FTAAS gateway (:8080)
┌─────────┼──────────┐
▼ ▼ ▼
registry control deploy
│
workflow
│
runner · Ray · MLflow · vLLM
Components
| Module | Role |
|---|---|
gateway (ftaas_app) |
Single process on :8080 — Console + APIs |
| console | Web UI for setup, tracking, prompt |
| registry | Dataset registration → id:version |
| control | Fine-tune jobs + model registry |
| workflow | Pipeline create / complete |
| deploy | Endpoints (vLLM / Ray Serve path) + prompt API |
| runner | Local (or Airflow) train pipeline |
| ftaas SDK | Client for notebooks / automation |
Quick start
python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
./scripts/start_all.sh # → http://127.0.0.1:8080
./scripts/e2e_smoke.sh
./scripts/stop_all.sh
SDK
from ftaas import Client, Framework, Technique, HyperParameters
with Client() as c:
ds = c.register_dataset("examples/data/alpaca_sample.jsonl", name="alpaca")
job = c.create_finetune_job(
model_name="sshleifer/tiny-gpt2",
dataset=ds,
framework=Framework.TRANSFORMERS,
technique=Technique.LORA,
parameters=HyperParameters(max_steps=10),
)
job = c.wait_for_job(job.job_id)
model = c.get_model(job.registered_model_name)
ep = c.create_endpoint(model.model_name, inference_framework="vllm")
print(c.prompt(ep.endpoint_id, "What is LoRA?").completion)
Layout
FTAAS/
├── packages/ftaas_sdk/ftaas/ # Client SDK
├── services/
│ ├── ftaas_app/ # gateway
│ ├── registry/
│ ├── control/
│ ├── workflow/
│ └── deploy/
├── runner/ # local + Airflow DAG
├── training/
├── ui/console/
├── examples/
└── scripts/
License
Apache-2.0
Metadata
Release files for ftaas 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| ftaas-0.1.0.tar.gz | 13.9 kB | Without provenance | Details for ftaas-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| ftaas-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for ftaas-0.1.0-py3-none-any.whl |
Total release size: 27.5 kB
Release files / ftaas-0.1.0.tar.gz
| Download URL | ftaas-0.1.0.tar.gz |
|---|---|
| Size | 13.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
d292d4ec337e9a69cb04d1d1f4877029ab93dba5f8d30eae9837b7cf0daebfa6
|
|
BLAKE2b-256 checksum How to use checksums |
1cc13df74227f226b073e159c28e9d1e9848a37f2afb80ba27ef281c0cb796bb
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.6
|
Release files / ftaas-0.1.0-py3-none-any.whl
| Download URL | ftaas-0.1.0-py3-none-any.whl |
|---|---|
| Size | 13.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
b556ac8ff68a51711b9e45f2a8bac9ac74dbfe649faa92085b268dcd0f21a24f
|
|
BLAKE2b-256 checksum How to use checksums |
d5ff03f2ababf0284c8788fd28c5fcd4c371b9a8fe3d9f5fb94eb0c3c07565ae
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.6
|
Security for ftaas 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.