Pareto Inference Provider Guide

Verification status: this entry is docs- and roster-verified only — not yet live-verified. It was onboarded credential-free: no Pareto Inference API key was ever created or used. Every field below comes from Pareto's public docs (https://docs.paretoinference.com/) and an unauthenticated GET /v1/models call (no auth header). No POST request was ever sent, so no generate, stream, tool-calling or structured-output capability has been exercised end to end. Treat capabilities.tools and .thinking as documentation-grade, not wire-proven, until evidence.liveMatrix is filled in by a live run.

Pareto Inference is a Tier-2 catalog provider: OpenAI-wire-compatible with no behavioural quirks, so its entire integration is one JSON file (src/lib/providers/catalog/pareto-inference.json) rather than hand-written code. That file is the source of truth for everything on this page.

---

Key Facts

• Provider id: pareto-inference
• Protocol: OpenAI-compatible (/chat/completions) — "Pareto accepts only OpenAI Chat Completions."
• Base URL: https://api.paretoinference.com/v1
• Default model: z-ai/glm-5.3-flash (GLM 5.3 Flash; the docs also document a bare alias, glm-5.3-flash, for the same model)
• Models in catalog: 1 — the only model Pareto serves. Confirmed by an unauthenticated GET /v1/models (no key needed), which returns exactly one id.
• Streaming: documented (SSE, text in choices[0].delta.content, tool calls in choices[0].delta.tool_calls, terminal data: [DONE]) — not live-probed.
• Tool calling: documented (tools, tool_choice) but the vendor's own request-parameter table hedges it with "Model support can differ," with no model-specific confirmation — set to model-dependent, not true.
• Structured output: response_format is a documented parameter but carries the same "Model support can differ" hedge as tools; the catalog schema has no model-dependent option for this field, so it is set false rather than asserted from a hedged table. structuredOutputWithTools is false — no combined probe was possible credential-free.
• Reasoning: reasoning_effort ("Sets the reasoning level") is documented as a plain, unhedged request parameter — unlike tools / response_format / tool_choice / logprobs, it carries no "Model support can differ" caveat — the basis for capabilities.thinking: true.
• Vision: not documented anywhere on the site.
• Embeddings: not documented anywhere on the site.
• Billing: no free tier. A payment card and prepaid credits are required before a key is issued.
• Signup: https://paretoinference.com/dashboard
• Key format: none declared.
• Data handling: Pareto documents zero data retention (ZDR) — request content is not kept after a request finishes.

---

Quick Start

1. Get an API key

1. Visit: https://paretoinference.com/dashboard and sign in
2. Add a payment card — Pareto has no free tier; a card and prepaid credits are required before a key is issued (https://docs.paretoinference.com/pricing)
3. Buy prepaid credits, then create your API key from the dashboard
4. Set PARETO_INFERENCE_API_KEY in your .env file

2. Configure

export PARETO_INFERENCE_API_KEY=your-api-key
export PARETO_INFERENCE_MODEL=z-ai/glm-5.3-flash      # optional — overrides the default model
export PARETO_INFERENCE_BASE_URL=https://api.paretoinference.com/v1   # optional — self-hosted or proxy

3. Use it

import { NeuroLink } from "@juspay/neurolink";

const neurolink = new NeuroLink();

const result = await neurolink.generate({
  input: { text: "Explain context windows in one paragraph." },
  provider: "pareto-inference",
  model: "z-ai/glm-5.3-flash",
});

console.log(result.content);

# CLI
npx @juspay/neurolink generate "Hello" --provider pareto-inference

Per-request credentials work as they do for every provider:

await neurolink.generate({
  input: { text: "Hello" },
  provider: "pareto-inference",
  credentials: {
    "pareto-inference": { apiKey: process.env.PARETO_INFERENCE_API_KEY },
  },
});

---

Models

Model	Context	Vision	$/M in · out · cached	Notes
z-ai/glm-5.3-flash ⭐	131,072*	no	$0.03 / $0.10 (cached $0.006)	Only model Pareto serves. Max output documented as 1–131,072 tokens (default 131,072). *Context window is not published by Pareto — this figure reuses the documented output ceiling as a conservative floor, not a stated context length.

* Pareto's own docs state it does not disclose GLM 5.3 Flash's serving details ("We do not disclose the technical details of our GLM 5.3 Flash serving system"), and no context-window number appears on any public page. models.defaultContextWindow (131,072) is therefore the documented output ceiling reused as a floor, flagged in the catalog entry's description — correct it once Pareto publishes a real figure or an authenticated probe reveals one.

Fallback order: none — z-ai/glm-5.3-flash is the only model Pareto serves, so it is both the default and its own (schema-exempt) single-entry fallback list.

---

Verification status

Tier-2 onboarding requires evidence before a provider is accepted, and pnpm run verify:provider-onboarding gates it in CI. This is what the catalog records for Pareto Inference — and, just as importantly, what it does not yet record:

Probe	Result
Roster	unauthenticated GET /v1/models, HTTP 200, 1 model (z-ai/glm-5.3-flash), 2026-09-28. No API key was used or required for this call.
Auth rejection	Not probed. Would require a POST with an invalid key, which this credential-free onboarding pass does not send. Pareto's docs describe 401 as "the chat API key is missing, invalid, or no longer active," with no example error body.
Live capability sweep	Not run. evidence.liveMatrix is null. capabilities.tools and .thinking are set from the vendor's own request-parameter table and its hedging language, not from an executed chat/stream/tool-call/schema request. structuredOutput and toolsWithStreaming are false because the same table hedges response_format and streamed tool calls with "Model support can differ" too.

Until a live run fills in evidence.liveMatrix, treat this provider as documentation-grade: the wire shape (base URL, auth header, request/ response field names) is well-documented and consistent with the standard OpenAI-compatible contract, but no NeuroLink call has actually exercised it.

---

Troubleshooting

Symptom	Cause	Fix
Invalid Pareto Inference API key	PARETO_INFERENCE_API_KEY unset, wrong, or expired	Pareto documents 401 as "the chat API key is missing, invalid, or no longer active" — check or replace the key in the dashboard
429 with credit_exhausted	Prepaid balance is empty	Buy more prepaid credits
429 with credit_insufficient	Held credits (reserved for max_tokens) are below what's needed	Lower max_tokens, or buy more credits
429 with model_capacity	The model itself is at capacity — not an account-level limit	Honor the Retry-After header and retry
502 mid-stream	The model request failed after the stream started	Retry the request; check whether a tool call already ran before retrying
503	Temporary Pareto-side failure (e.g. maintenance, GPU outage)	Wait and retry; Retry-After gives the delay in seconds
No context-window figure in the catalog	Pareto does not publish one and declines to disclose serving details	Treat defaultContextWindow as a conservative floor, not a hard vendor number, until corrected by a real probe

---

See also

• Provider setup overview
• All providers
• Tier-2 onboarding — how this provider's JSON becomes a working integration
• Provider feature compatibility