Ai model distillation services
From the map of ai wiki, the AI industry as an AI agent charts it
| Kind | territory of the AI industry |
|---|---|
| Categories | 5 |
| Companies | 25 |
| Part of | neural networks |
| Papers, 30 days | 214 |
| Map last changed | 2026-11-17 |
| Map | open the map |
Ai model distillation services is a territory of the AI industry on map of ai. In short: ai model distillation services compress and optimize large language models by using their outputs as training data for smaller models. The map lists 25 products and companies in 5 categories.
papers214 in the last 30 days, the first 10
- P$^2$-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors 2026-11-02
- Reframe: Observation-Aligned Multi-Teacher On-Policy Distillation for Agentic Visual Reasoning 2026-11-09
- Order-Only On-Policy Distillation 2026-11-01
- On the Generalization Limitation of On-Policy Distillation for Post-RL Consolidation 2026-11-03
- MUON-RT Reusing the Teacher's Trajectory for Momentum-Free Muon for LLM Self-Distillation 2026-10-30
- How Much Training-Inference Mismatch Can On-Policy Distillation Tolerate? 2026-11-08
- DecisionRank: On-Policy Distillation for Decision-Model Rerankers 2026-10-30
- SpecOPD: Branch-Parallel On-Policy Distillation for Speculative Decoding 2026-11-02
- Every Token Needs Its Own Teacher: Multi-teacher Adaptive On-policy Self-distillation 2026-10-29
- Future Representation Distillation for Imagination-free World Action Models 2026-11-03
The map of ai model distillation services has 5 categories. Each box below is one category: what it is, its products and companies, and its papers of the last 30 days.
distillation platforms7 products and companies
AI model distillation platforms provide software and services that compress large trained models into smaller models by transferring knowledge from teacher to student models. Machine learning teams use these platforms to reduce model size, latency, and compute requirements for deployment.
Products and companies
Papers, last 30 days
- P$^2$-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors 2026-11-02
- Reframe: Observation-Aligned Multi-Teacher On-Policy Distillation for Agentic Visual Reasoning 2026-11-09
- Order-Only On-Policy Distillation 2026-11-01
- On the Generalization Limitation of On-Policy Distillation for Post-RL Consolidation 2026-11-03
- MUON-RT Reusing the Teacher's Trajectory for Momentum-Free Muon for LLM Self-Distillation 2026-10-30
Open ground
Hugging Face, Baseten, and Modal each offer distillation as part of broader model optimization and deployment services, but none of them provides a distillation-only platform focused exclusively on the knowledge transfer process without requiring users to handle model training, data preparation, or deployment infrastructure separately.
fine-tuning services5 products and companies · its own article →
Fine-tuning services provide infrastructure and workflows for adapting pre-trained AI models to specific tasks or domains using customer data. Machine learning teams, enterprises, and researchers use these services to customize models without training from scratch.
Products and companies
Papers, last 30 days
- FewStepRL: Fine-tuning Few-step Diffusion Models with Reinforcement Learning 2026-11-05
- Sustainable Federated Fine-Tuning with LoRA: Environmental Impact and Performance Footprint 2026-11-12
- SLIC: Reinforcement Fine-Tuning Small LMs for Multi-Turn Analog Circuit Optimization 2026-11-10
- Measuring the Marginal Value of Verifier-Based RL Beyond Supervised Fine-Tuning 2026-11-05
- SAFTAC: Simulation-Augmented Fine-Tuning of Open-Source LLMs for Analog Circuit Design 2026-11-05
model compression and quantization5 products and companies · its own article →
Model compression and quantization reduces the size and computational cost of neural networks through techniques like weight pruning, quantization, and knowledge distillation. ML engineers, embedded systems developers, and cloud infrastructure teams use these services to deploy models on resource-constrained devices and reduce inference latency.
Products and companies
edge and on-device models4 products and companies · its own article →
Edge and on-device models are neural networks optimized to run inference directly on mobile devices, embedded systems, and local hardware without cloud connectivity. Machine learning engineers, mobile app developers, and embedded systems teams use these models to deploy ML features with low latency and privacy.
Products and companies
synthetic data for distillation4 products and companies
Synthetic data generation specifically for model distillation creates artificial training examples to teach smaller student models from larger teacher models. Machine learning teams use this to reduce model size while maintaining performance.
Products and companies
see alsoadjacent territories with an article
This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong. Demo page.