◎ Discry Score
together ai.com
ai · API
A
0 / 100
DISCOVERY0
COMPREHENSION0
Category leader: 99 (A)
Discry your API →
AI · RANK #18 OF 43

Together AI is Agent-Ready to agents.

Discry independently scored how well an AI agent can discover and understand the Together AI API from what’s public — not whether it’s usable. Below: every signal we checked, what’s costing the score, and what to change.

Discry your API — freeView the docs ↗

SCORED UNDER RUBRIC 1.2 · A full re-launch under Discry Score 2.5 — a new behavioral instrument, not comparable to these scores — is in progress.

Discovery

45% of score · 86/100
OpenAPI specA machine-readable OpenAPI/Swagger spec agents can parse.Pass
llms.txtAn llms.txt index that points agents to the docs that matter.Pass
llms.txt qualityThe llms.txt is focused, current, and well under the size budget.Pass
llms-full.txtA full-text bundle agents can load in one request.Pass
AGENTS.mdAn AGENTS.md that tells coding agents how to build on the API.Fail
.well-known/mcp.jsonA discoverable MCP manifest at a well-known path.Pass
MCP registryThe API is listed in a public MCP registry.Pass
robots.txt AI directivesrobots.txt allows (or explicitly guides) AI crawlers.Pass
SitemapA sitemap so agents can enumerate the docs surface.Pass

Comprehension

55% of score · 96/100
Task-oriented descriptionsEndpoints described by what they accomplish, not just their shape.Pass
Realistic examplesRunnable, real-world request/response examples.Pass
Multi-step workflowsDocs that chain calls into complete jobs an agent can follow.Pass
Error-recovery guidanceDocumented failure modes and how to recover from them.Pass
Answer-first formatThe answer leads; preamble does not bury it.Pass
Capability boundariesClear limits — what the API can and cannot do.Pass
Naming consistencyConsistent, predictable naming across endpoints.Pass
Heading hierarchyClean heading structure agents can navigate.Pass
Markdown docsDocs available as clean markdown, not JS-rendered HTML only.Pass
Token efficiencyDocs are concise enough to fit an agent context window.Partial

What we found

  • An agent discovering Together AI finds a near-complete readiness surface: robots.txt explicitly opts into AI training and input (Content-Signal: ai-train=yes, ai-input=yes), a 42KB API-focused llms.txt index, a 1.8MB llms-full.txt, a valid OpenAPI 3.1 spec with embedded code samples, and a live .well-known/mcp.json pointing to a docs MCP server.
  • Error recovery is exemplary — every status code from 400 to 529 has an explicit cause and fix, and rate-limit docs prescribe exponential backoff keyed to the x-ratelimit-reset header. An agent hitting a 429, 403, or 402 is told exactly what to change rather than left to guess.
  • Every documentation page is served as clean markdown (append .md to any URL) with multi-language, copy-pasteable code samples (Python/TypeScript/cURL) and OpenAI-compatible snake_case naming, so an agent can parse and execute without HTML wrangling.
  • The main discovery gap is the absence of an AGENTS.md in any primary togethercomputer repo (together-python, together-typescript, openapi, MoA all 404) — a coding agent cloning the SDK gets no repo-level instructions, though the docs do ship 'agent skills' and a live MCP server as an alternative.
  • Dozens of end-to-end workflow guides (RAG pipeline, AI search engine, step-by-step batch jobs, conditional workflows) mean an agent can chain operations for real tasks, not just call endpoints in isolation.

What to change

Prioritized by impact on discoverability. You (or your docs platform) deploy these — Discry never touches your API.

  1. 01Add an AGENTS.md to the primary SDK repos (together-python, together-typescript, openapi) — the single highest-impact discovery fix. At weight 3 it would lift the discovery score from 86 to ~100.
  2. 02Document idempotency keys (or explicitly state which write endpoints are non-idempotent and their safe-retry semantics) so agents can retry create/upload calls without risking duplicate side effects.
  3. 03Publish a tighter 'core capabilities' overview, or trim the 42KB llms.txt index, so an agent can grasp the primary inference surface in under 5,000 tokens without parsing the full index.
  4. 04Pursue an official, first-party MCP server listing in the major registries (Glama, Smithery, PulseMCP currently surface only third-party image-generation wrappers) so the canonical Together MCP is the top discovery hit.

Execution coverage · INFORMATIONAL, UNSCORED

Whether an agent can actually complete a call and recover from errors is the deeper Audit layer — documented here, but not part of the Discry Score.

Bearer API key Error format documented Rate limits documented Pagination documented Idempotency documented

Execution characteristics are documented thoroughly. Auth is a single Bearer API key in the Authorization header (account signup uses Google/GitHub OAuth, but the API itself is key-based). Errors return OpenAI-shaped JSON objects ({error:{message,type,code}}) with a full status-code table. Rate limits are dynamic per-model, surfaced via 429 plus an x-ratelimit-reset header with exponential-backoff guidance. Pagination is cursor-based (after/--after). Idempotency keys are not documented — retry safety is described informally for transient 5xx errors only.

See your own Discry Score.

Drop your API docs URL. See what an agent sees — in 60 seconds, free.

Discry your API — free