# Adaptive ML

**Canonical:** https://apis.io/providers/adaptive-ml/  
**Website:** https://www.adaptive-ml.com  
**APIs profiled:** 1

Adaptive ML builds Adaptive Engine, an enterprise platform for developing, evaluating, and serving specialized open-source large language models via reinforcement-learning post-training. The three-phase workflow (Adapt, Evaluate, Serve) lets teams fine-tune smaller models to outperform commercial APIs, measure them with custom AI judges and graders, and feed production signals back into training. Adaptive Engine is self-hosted (Kubernetes/Helm) and exposes an OpenAI-compatible REST API for chat completions and embeddings, plus interactions, comparisons, outcomes, dataset/recipe management, and chunked uploads. Tooling includes the Python adaptive-sdk and adaptive-harmony libraries and the adpt CLI. Adaptive ML was acquired by Datadog.

## Kin Score — 41.5 / 100 (developing)

Scored 2026-08-30 under rubric 0.17.2. Trend: flat (-0.6 from 42.1).

| Facet | Score |
|---|---|
| Discoverability | 75.9 |
| Contract Quality | 51.8 |
| Governance | 4.5 |
| Contract Governance | 4.5 |
| Operational Transparency | 26.3 |
| Developer Ergonomics | 61.3 |
| Commercial Clarity | 23.7 |
| Access Clarity | 23.7 |

## Agent readiness — 25.4 (agent-aware)

| Dimension | Value |
|---|---|
| Spec Presence | yes |
| Agentic Access | derived |
| Reversibility Documented | no |
| MCP Server | no |
| Auth Clarity | bearer |
| Idempotency | no |
| Error Semantics | documented |
| OpenAPI Examples | no |
| Rate Limit Signal | no |
| Event Surface Described | yes |
| Agent Skills | derived |
| Well Known Catalog | no |
| Consent Identity | no |
| Agent Card | no |
| Dry Run Mode | no |
| Delegated Identity | no |
| Protected Resource Metadata | no |
| Dynamic Client Registration | no |
| Agentic Commerce | no |

## Access

Self-serve signup — onboarding: self-serve, pricing: unknown, trial: no (confidence: medium).

## APIs (9)

- **Adaptive ML artifacts::rest API** — The artifacts::rest API from Adaptive ML — 1 operation(s) for artifacts::rest.
- **Adaptive ML Chunked Upload API** — Upload large files in chunks
- **Adaptive ML Completions API** — The Completions API from Adaptive ML — 1 operation(s) for completions.
- **Adaptive ML Datasets API** — The Datasets API from Adaptive ML — 1 operation(s) for datasets.
- **Adaptive ML Embeddings API** — The Embeddings API from Adaptive ML — 1 operation(s) for embeddings.
- **Adaptive ML Feedback API** — The Feedback API from Adaptive ML — 2 operation(s) for feedback.
- **Adaptive ML image::rest API** — The image::rest API from Adaptive ML — 1 operation(s) for image::rest.
- **Adaptive ML Interactions API** — Load interactions in the db
- **Adaptive ML Recipes API** — Recipe operations

## MCP servers (1)

- **Adaptive ML MCP Server**

## Agentic access (1)

- **Adaptive Ml Agentic Access** — 13 operations · 9 acting

## Security (2)

- **Adaptive Ml Authentication** — http · 1 scheme
- **Adaptive Ml Domain Security** — TLSv1.3 · HSTS · DMARC

## Tags

Company, Ai Ml, LLM, Fine-Tuning, Reinforcement Learning, Model Evaluation, Inference, OpenAI-Compatible, MLOps

---

Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/adaptive-ml/). Scores are computed from the provider's own public artifacts under a published rubric.