Loops
Python SDK for LoRA fine-tuning with supervised fine-tuning and reinforcement learning
- Upvotes
- 0
- DR
- 77 no change since last check
- Followers
- 0
Python SDK for LoRA fine-tuning and reinforcement learning on managed training infrastructure from Baseten.
About Loops
Loops is a Python SDK for LoRA fine-tuning that supports both supervised fine-tuning (SFT) and reinforcement learning (RL) on managed training and sampling infrastructure. You write the training script and Baseten runs the trainer and sampler. The SDK provides four core operations: forward_backward() for computing loss and accumulating gradients, optim_step() for updating the adapter, sample() for generating from the current or saved policy, and save_state() for saving training state. The API is compatible with Tinker. Training and sampling run on separate resources, with options for serverless shared capacity or dedicated GPUs. Loops trains LoRA adapters on open models from publishers including Qwen, Google, DeepSeek, Moonshot AI, NVIDIA, and Z.ai, including vision-language models.
- lora-finetuning
- supervised-finetuning
- reinforcement learning
- training infrastructure
- model training
- machine-learning
- python
Features worth knowing
- forward_backward() operation for computing loss and accumulating gradients
- optim_step() for updating the adapter
- sample() for generating from current or saved policy
- save_state() to save and resume training state
- Tinker API compatibility
- Separate trainer and sampler resources
- Synchronous and asynchronous RL patterns
- Frozen base model weights with LoRA adapter updates
- Support for open models from Qwen, Google, DeepSeek, Moonshot AI, NVIDIA, and Z.ai
Follow Loops and hear about their milestones, new launches and demo videos by email. Stop any time.
Loops metrics
| Metric | Value |
|---|---|
| Domain Rating | 77 no change since last check |
| Trust Flow | 13 no change since last check |
| Citation Flow | 43 no change since last check |
| Referring domains | 550 no change since last check |
| External backlinks | 3,862 no change since last check |
Ahrefs Domain Rating scanned 11 Nov 2026, updated 8 Nov 2026. Main topic: Society/Subcultures (13). Majestic scanned 8 Nov 2026, updated 8 Nov 2026. Link stats via Majestic. Trust Flow, Citation Flow and Topical Trust Flow are trademarks of Majestic-12 Ltd.
Loops domain facts
Reveal the domain facts
No RDAP data is published for .co, so this domain's registration details are not available here.
Loops launch results
| Measure | At launch, then latest |
|---|---|
| Domain Rating (Ahrefs) | 77 to 77 no change since launch 8 Nov 2026, then 10 Nov 2026 |
| Trust Flow (Majestic) | 13 first measured 8 Nov 2026 |
| Referring domains (Majestic) | 550 first measured 8 Nov 2026 |
Day 2 since launch. Next receipt: day 30, 8 Dec 2026. The figures report a change; they do not say what caused it, and many things can. No figure here is a guess.
Launching something soon?
Free tips from Best Digital Launch for getting ready, launch day, the week after and what to do next.
- Set up
- On launch day
- The days after
- Beyond launch
Compare and explore
FAQ
- What are the four core operations of the Loops SDK?
- The SDK's four core operations are forward_backward() which computes the loss and accumulates gradients, optim_step() which updates the adapter, sample() which generates from the current or saved policy, and save_state() which saves training state that can be resumed.
- How do training and sampling work in Loops?
- Your training script sends each batch to the trainer, which computes the loss and gradients and updates the LoRA adapter while the base model's weights stay frozen. To generate with the updated adapter, the script publishes the adapter's weights and samples from them. Training and sampling run on separate resources.
- Which models does Loops support?
- Loops trains LoRA adapters on open models from publishers including Qwen, Google, DeepSeek, Moonshot AI, NVIDIA, and Z.ai, including vision-language models.
Leave a review for Loops
Nobody has reviewed Loops yet. Start things off.
A review is always free and can never be paid for. Go to the reviews and discussion
Reviews and chat · Product news · Send it on · Run this company? Claim it here · Report it
Loops has a free listing. The owner can unlock the full profile with a paid launch for $5:
- Founder story (locked)
- Milestones (locked)
- Demo video (locked)
- Dofollow website link (locked)
Alternatives to Loops
YABANIN
Self-hosted AI infrastructure with measurable returns
AI inference & GPU cloud0 votes. Log in to add your vote for YABANINQuantum Computing Inc
Putting photons to work
AI inference & GPU cloud0 votes. Log in to add your vote for Quantum Computing IncKinetic
Open-source engine for faster local language model inference with bit-for-bit identical output
AI inference & GPU cloud0 votes. Log in to add your vote for KineticKaeso
Open-source AI gateway you run yourself
AI inference & GPU cloud0 votes. Log in to add your vote for Kaeso
