Positron

Positron AI is a US-based generative-AI hardware company founded in 2023 and headquartered in Reno, Nevada, building purpose-built Transformer inference accelerators as an alternative to GPU-based inference infrastructure. Its shipping product, Atlas, is a rack inference appliance built on eight in-house Archer accelerators (256 GB HBM) and dual AMD EPYC Genoa processors, with the next-generation Asimov silicon and Titan system slated for 2027. Positron exposes its inference fleet through "Olivaw", a serving and administration layer that publishes two documented HTTP APIs: an OpenAI-compatible completions API (models, chat completions, text completions, with server-sent-event streaming) so existing OpenAI client code can be repointed at a Positron endpoint, and an administrative API for managing models, service nodes, users, and access tokens on an appliance or hosted cluster. The company raised a $230M Series B in February 2026 at a valuation above $1B from ARENA, Jump Trading, Unless, the Qatar Investment Authority, Arm, Helena, DFJ Growth, Atreides, Valor, Resilience Reserve, Flume Ventures and 1517 Fund.

Positron publishes 2 APIs on the APIs.io network: Olivaw OpenAI-Compatible Inference API and Olivaw Admin API. Tagged areas include artificial-intelligence, ai-inference, inference-hardware, ai-accelerators, and large-language-models.

Positron’s developer surface includes documentation, API reference, getting-started guide, support, changelog, authentication, and 23 more developer resources.

🌐 Visit website 📡 Source on GitHub
37.0/100 thin ▬ flat Agent 39/100 agent ready Full breakdown ↓
scored 2026-08-05 · rubric v0.9.1
2 APIs
artificial-intelligenceai-inferenceinference-hardwareai-acceleratorslarge-language-modelstransformersopenai-compatiblemachine-learningsemiconductorsdata-centerllm-servingmodel-hosting

Kin Score

Kin Score Kin Score How this is scored →
scored 2026-08-05 · rubric v0.9.1
Composite quality — 37.0/100 · thin
Contract Quality 13.9 / 25
Developer Ergonomics 10.3 / 20
Commercial Clarity 0.0 / 20
Operational Transparency 2.7 / 13
Governance 1.4 / 12
Discoverability 8.7 / 10
Agent readiness — 39/100 · agent ready
Machine-Readable Contract 18 / 18
Agentic Access Contract 10 / 10
MCP Server 0 / 12
Machine-Readable Auth 10 / 10
Idempotency 0 / 9
Stable Error Semantics 8 / 8
Request/Response Examples 0 / 7
Rate-Limit Signaling 0 / 7
Typed Event Surface 0 / 6
Agent Skills 5 / 5
Well-Known Catalog 4 / 4
Consent & Bot Identity 0 / 3
A2A Agent Card 0 / 8
Dry-Run / Simulate Mode 0 / 4
Improve this rating by publishing the missing artifacts — every area above can be raised, and the full rubric is at apis.io/rating/. This rating is computed from github.com/api-evangelist/positron: open an issue to ask a question, or submit a pull request to add artifacts. Want it done for you? Prioritized profiling — $2,500 →

APIs 2

Individual APIs this provider publishes, each with its own machine-readable definition.

Positron Olivaw OpenAI-Compatible Inference API

OpenAI API-compatible inference endpoint served by Positron's Olivaw serving layer. Lists available models, retrieves a single model, and creates chat completions and raw text c...

Positron Olivaw Admin API

Administrative API for the Olivaw layer that fronts Positron Atlas hardware. Manages the model catalog (list/create/read/update/delete), the service nodes that back it (Giskard,...

Security Posture 2

Authentication, domain security, vulnerability disclosure, and trust-center signals.

Positron Authentication

apiKey/http-bearer · 2 schemes

SECURITY

Positron Domain Security

TLSv1.3 · HSTS · DMARC

SECURITY

Agentic Access 1

Recommended x-agentic-access execution contracts for AI agents.

Positron Agentic Access

23 operations · 13 acting

23 operations · 13 acting

AGENTIC

Resources

Get Started 2

Portal, sign-up, and the first successful call

DeveloperPortal
DeveloperPortal
 GettingStarted
GettingStarted

Documentation 4

Reference material describing how the API behaves

Documentation
Documentation
 APIReference
APIReference
 OpenAPI
OpenAPI
 OpenAPI
OpenAPI

Agent Surfaces 4

MCP servers, agent skills, and machine-readable catalogs

AgenticAccess
AgenticAccess
 WellKnown
WellKnown
 LLMsTxt
LLMsTxt
 AgentSkill
AgentSkill

Design & Contract 5

Pagination, idempotency, versioning, errors, and events

Conventions
Conventions
 ErrorCatalog
ErrorCatalog
 DataModel
DataModel
 Lifecycle
Lifecycle
 Conformance
Conformance

Build 2

SDKs, sample code, and the tooling you integrate with

GitHubOrganization
GitHubOrganization
 Packages
Packages

Access & Security 2

Authentication, authorization, and security posture

DomainSecurity
DomainSecurity
 Authentication
Authentication

Operate 3

Status, limits, changes, and where to get help

Support
Support
 ChangeLog
ChangeLog
 ChangeLog
ChangeLog

Company 5

The organization behind the API

Website
Website
 Press
Press
 Careers
Careers
 LinkedIn
LinkedIn
 Twitter
Twitter

Other 2

Properties that don't map to a standard resource type

Overlay
Overlay
 Overlay
Overlay

Source (apis.yml)

apis.yml Raw ↑ 
aid: positron
name: Positron
description: 'Positron AI is a US-based generative-AI hardware company founded in 2023 and headquartered in Reno, Nevada,
  building purpose-built Transformer inference accelerators as an alternative to GPU-based inference infrastructure. Its shipping
  product, Atlas, is a rack inference appliance built on eight in-house Archer accelerators (256 GB HBM) and dual AMD EPYC
  Genoa processors, with the next-generation Asimov silicon and Titan system slated for 2027. Positron exposes its inference
  fleet through "Olivaw", a serving and administration layer that publishes two documented HTTP APIs: an OpenAI-compatible
  completions API (models, chat completions, text completions, with server-sent-event streaming) so existing OpenAI client
  code can be repointed at a Positron endpoint, and an administrative API for managing models, service nodes, users, and access
  tokens on an appliance or hosted cluster. The company raised a $230M Series B in February 2026 at a valuation above $1B
  from ARENA, Jump Trading, Unless, the Qatar Investment Authority, Arm, Helena, DFJ Growth, Atreides, Valor, Resilience Reserve,
  Flume Ventures and 1517 Fund.'
image: https://www.positron.ai/opengraph-image-pwu6ef.png
url: https://raw.githubusercontent.com/api-evangelist/positron/refs/heads/main/apis.yml
x-type: company
x-source: harvest:secondary-market
specificationVersion: '0.21'
created: '2026-08-02'
modified: '2026-08-02'
tags:
- artificial-intelligence
- ai-inference
- inference-hardware
- ai-accelerators
- large-language-models
- transformers
- openai-compatible
- machine-learning
- semiconductors
- data-center
- llm-serving
- model-hosting
apis:
- name: Positron Olivaw OpenAI-Compatible Inference API
  description: OpenAI API-compatible inference endpoint served by Positron's Olivaw serving layer. Lists available models,
    retrieves a single model, and creates chat completions and raw text completions against models loaded onto Positron Atlas
    hardware, with optional text/event-stream streaming responses and Positron-specific selection settings (max_tokens, temperature,
    top_p, top_k). Authenticated with an API key in the authorization header. Because the API mirrors the OpenAI surface,
    existing OpenAI client libraries can be repointed at a Positron endpoint without code changes.
  humanURL: https://support.positron.ai/api-documentation
  baseURL: https://api.positron.ai/
  x-baseurl-evidence: api.positron.ai is live (Caddy, HTTP/2) and returns 401 Unauthorized on every probed path, including
    /openapi.json, /v1/models and /.well-known/*, so the surface is entirely auth-gated and could not be enumerated anonymously
    (probed 2026-08-02). The published OpenAPI declares a deployment-scoped server template `{host}/api/v1` whose default
    host is `http://localhost:4000` (an on-premises Atlas appliance); the user guide documents hosted endpoints in the form
    `https://<groupname>-olivaw.fly.dev`.
  tags:
  - ai-inference
  - large-language-models
  - openai-compatible
  - llm-serving
  - chat-completions
  properties:
  - type: OpenAPI
    url: openapi/positron-inference-openapi.yml
  - type: Documentation
    url: https://support.positron.ai/api-documentation
  - type: APIReference
    url: https://support.positron.ai/api-documentation
  - type: Authentication
    url: authentication/positron-authentication.yml
  - type: Conventions
    url: conventions/positron-conventions.yml
  - type: ErrorCatalog
    url: errors/positron-problem-types.yml
  - type: DataModel
    url: data-model/positron-data-model.yml
  - type: Overlay
    url: overlays/positron-inference-overlay.yaml
- name: Positron Olivaw Admin API
  description: Administrative API for the Olivaw layer that fronts Positron Atlas hardware. Manages the model catalog (list/create/read/update/delete),
    the service nodes that back it (Giskard, vLLM or OpenAI backends, with per-node JSON configuration and model lists), platform
    users, and API access tokens. Published by Positron AI as an OpenAPI 3.1.0 document in the public positron-ai/admin-api-docs
    repository; secured with a bearer API key.
  humanURL: https://github.com/positron-ai/admin-api-docs
  baseURL: https://api.positron.ai/
  x-baseurl-evidence: The published admin OpenAPI declares no servers[] block; the admin surface is served per deployment
    by the Olivaw appliance/cluster. api.positron.ai is Positron's live, auth-gated (401) API host as probed 2026-08-02.
  tags:
  - ai-inference
  - model-hosting
  - administration
  - access-tokens
  - llm-serving
  properties:
  - type: OpenAPI
    url: openapi/positron-admin-openapi.yml
  - type: Documentation
    url: https://github.com/positron-ai/admin-api-docs
  - type: SourceCode
    url: https://github.com/positron-ai/admin-api-docs
  - type: Authentication
    url: authentication/positron-authentication.yml
  - type: ErrorCatalog
    url: errors/positron-problem-types.yml
  - type: DataModel
    url: data-model/positron-data-model.yml
  - type: Overlay
    url: overlays/positron-admin-overlay.yaml
common:
- type: DomainSecurity
  url: security/positron-domain-security.yml
- type: AgenticAccess
  url: agentic-access/positron-agentic-access.yml
- type: Website
  url: https://www.positron.ai/
- type: DeveloperPortal
  url: https://support.positron.ai/
- type: Documentation
  url: https://support.positron.ai/api-documentation
- type: APIReference
  url: https://support.positron.ai/api-documentation
- type: GettingStarted
  url: https://support.positron.ai/user-guide
- type: Support
  url: https://www.positron.ai/contact-sales
- type: GitHubOrganization
  url: https://github.com/positron-ai
- type: ChangeLog
  url: https://support.positron.ai/release-notes
- type: ChangeLog
  url: changelog/positron-changelog.yml
- type: Press
  url: https://www.positron.ai/press
- type: Careers
  url: https://www.positron.ai/careers
- type: LinkedIn
  url: https://www.linkedin.com/company/positron-ai/
- type: Twitter
  url: https://twitter.com/positron_ai
- type: OpenAPI
  url: openapi/positron-inference-openapi.yml
- type: OpenAPI
  url: openapi/positron-admin-openapi.yml
- type: Authentication
  url: authentication/positron-authentication.yml
- type: Conventions
  url: conventions/positron-conventions.yml
- type: ErrorCatalog
  url: errors/positron-problem-types.yml
- type: DataModel
  url: data-model/positron-data-model.yml
- type: Lifecycle
  url: lifecycle/positron-lifecycle.yml
- type: Conformance
  url: conformance/positron-conformance.yml
- type: Packages
  url: packages/positron-packages.yml
- type: WellKnown
  url: well-known/positron-well-known.yml
- type: LLMsTxt
  url: llms/positron-llms.txt
- type: Overlay
  url: overlays/positron-inference-overlay.yaml
- type: Overlay
  url: overlays/positron-admin-overlay.yaml
- type: AgentSkill
  url: skills/_index.yml
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com
- FN: APIs.json
  email: info@apis.io
- FN: Positron AI
  email: api@positron.ai
  url: https://www.positron.ai/
x-enrichment:
  date: '2026-08-02'
  status: enriched
  artifacts_added: 23
  pass: local-v1