From d7ecabee4211d304341004a427dcb1cbf21451dc Mon Sep 17 00:00:00 2001 From: Claude Date: Fri, 13 Nov 2026 04:38:58 +0000 Subject: [PATCH] docs: sync model catalog with dashboard API - gliner2.5-decide-v1: input $0.004 -> $0.01 (catalog table + card pill, text-classification table) - add d1.1: catalog row + card + Text Classification library card (8 -> 9 models), text-classification table row and `## d1.1` section, api-reference/models/d1-1.mdx, playgrounds/d1_1.openapi.json, System One page row and model list, docs.json nav - d1.1 is documented as System One only: the dashboard description states it is sent to POST /v1/systemone and does not chat, stream or call tools, so it is not added to the Responses or Chat enums Source: https://api-dashboard.zerogpu.ai/api/models Co-Authored-By: Claude Opus 5 Claude-Session: https://claude.ai/code/session_01V6BqNyDmLjxxtWuFoxFdXx --- api-reference/models/d1-1.mdx | 18 + .../openapi/playgrounds/d1_1.openapi.json | 972 ++++++++++++++++++ api-reference/openapi/zerogpu.openapi.json | 2 +- api-reference/systemone.mdx | 5 +- docs.json | 1 + docs/model-catalog.mdx | 17 +- docs/text-classification.mdx | 55 +- 7 files changed, 1063 insertions(+), 7 deletions(-) create mode 100644 api-reference/models/d1-1.mdx create mode 100644 api-reference/openapi/playgrounds/d1_1.openapi.json diff --git a/api-reference/models/d1-1.mdx b/api-reference/models/d1-1.mdx new file mode 100644 index 0000000..2be6387 --- /dev/null +++ b/api-reference/models/d1-1.mdx @@ -0,0 +1,18 @@ +--- +title: "d1.1" +sidebarTitle: "d1.1" +description: "Model details for d1.1. LiquidAI's decision model, answering named questions about a state on System One." +openapi: "api-reference/openapi/playgrounds/d1_1.openapi.json POST /systemone" +--- + + + This model is routable only on `/v1/systemone` (see + [System One](/api-reference/systemone)). Send the content to judge as + `state` — a string, an object, or an array — and up to 16 named `questions` + about it. It does not chat, stream, or call tools, so it 400s on + `/v1/responses` and `/v1/chat/completions`. + + +> LiquidAI's d1.1 is a decision model and the successor to d1. Send a state — text, JSON, or images — together with several named questions, and every question is answered in a single forward pass: `noul` returns the probability of yes, `choice` picks one of your named options with a confidence and the full probability distribution, and `score` rates the state on 2 to 10 ordered levels. It generates no text, so output tokens are always 0 and only the input is billed, across a 65,536-token state. It is built for intent classification, agent routing, moderation, commercial-intent detection, triage, and LLM-as-a-judge checks, and it does not chat, stream, or call tools. + +**References:** [Terms](https://www.liquid.ai/terms-conditions) • [Privacy](https://zerogpu.ai/privacy-policy) diff --git a/api-reference/openapi/playgrounds/d1_1.openapi.json b/api-reference/openapi/playgrounds/d1_1.openapi.json new file mode 100644 index 0000000..1477533 --- /dev/null +++ b/api-reference/openapi/playgrounds/d1_1.openapi.json @@ -0,0 +1,972 @@ +{ + "openapi": "3.1.0", + "info": { + "title": "d1.1 playground", + "version": "1.0", + "description": "Interactive playground for **d1.1**.\nModel is always `d1.1` on this page (shown in the form, not editable).\nd1.1 is routable only on `/v1/systemone` — it 400s on `/v1/responses` and `/v1/chat/completions`.\nSend the content to judge as `state` and up to 16 named `questions` about it; each answer comes back under its question's name and `output_tokens` is always 0.\nAuthentication: `x-api-key`, or the OpenAI-compatible `Authorization: Bearer` header. `x-project-id` is optional." + }, + "servers": [ + { + "url": "https://api.zerogpu.ai/v1", + "description": "Production" + } + ], + "security": [ + { + "ApiKey": [] + }, + { + "BearerAuth": [] + } + ], + "paths": { + "/systemone": { + "post": { + "operationId": "createSystemOneDecision_d1_1", + "summary": "d1.1: System One", + "description": "Answer named `noul`, `choice` and `score` questions about `state` in a single forward pass. No text is generated, so `output_tokens` is always 0 and only the input is billed.", + "tags": [ + "d1.1" + ], + "requestBody": { + "required": true, + "content": { + "application/json": { + "schema": { + "$ref": "#/components/schemas/SystemOneRequest" + }, + "example": { + "model": "d1.1", + "state": "Can you refund my last order? It arrived broken and I want to speak to someone.", + "questions": { + "handoff": { + "type": "noul", + "instructions": "Does this request need a human agent?" + }, + "intent": { + "type": "choice", + "instructions": "Which intent best describes this request?", + "criteria": { + "refund": "Asking for money back", + "payment_issue": "A payment failed or was charged twice", + "account_access": "Cannot sign in or recover an account", + "general_question": "Anything else" + } + }, + "urgency": { + "type": "score", + "instructions": "How urgent is this request?", + "criteria": [ + "None", + "Low", + "Medium", + "High" + ] + } + } + } + } + } + }, + "responses": { + "200": { + "description": "Success", + "content": { + "application/json": { + "schema": { + "$ref": "#/components/schemas/SystemOneResponse" + } + } + } + }, + "400": { + "description": "The model does not support System One (`model_not_supported`)", + "content": { + "application/json": { + "schema": { + "$ref": "#/components/schemas/ErrorResponse" + } + } + } + }, + "401": { + "description": "Unauthorized (missing API key)" + }, + "402": { + "description": "Insufficient quota (insufficient_quota)", + "content": { + "application/json": { + "schema": { + "$ref": "#/components/schemas/ErrorResponse" + } + } + } + }, + "404": { + "description": "Model not found (`model_not_found`)", + "content": { + "application/json": { + "schema": { + "$ref": "#/components/schemas/ErrorResponse" + } + } + } + }, + "422": { + "description": "Validation error", + "content": { + "application/json": { + "schema": { + "$ref": "#/components/schemas/HTTPValidationError" + } + } + } + }, + "502": { + "description": "Upstream inference failed (`upstream_error`)", + "content": { + "application/json": { + "schema": { + "$ref": "#/components/schemas/ErrorResponse" + } + } + } + } + } + } + } + }, + "components": { + "securitySchemes": { + "ApiKey": { + "type": "apiKey", + "in": "header", + "name": "x-api-key" + }, + "BearerAuth": { + "type": "http", + "scheme": "bearer", + "description": "OpenAI-compatible: send your ZeroGPU API key as `Authorization: Bearer `." + } + }, + "schemas": { + "CreateModerationRequest": { + "type": "object", + "required": [ + "model", + "input" + ], + "properties": { + "model": { + "type": "string", + "const": "zlm-v1-moderation-edge", + "default": "zlm-v1-moderation-edge", + "example": "zlm-v1-moderation-edge", + "description": "Model identifier (fixed for this playground)." + }, + "input": { + "description": "Text to classify. Accepts a single string, an array of strings (one result per element), or an array of content parts (`{ \"type\": \"text\", \"text\": \"...\" }`) that form a single multi-modal input.", + "oneOf": [ + { + "type": "string", + "minLength": 1, + "format": "textarea", + "maxLength": 131072 + }, + { + "type": "array", + "minItems": 1, + "items": { + "type": "string", + "minLength": 1, + "maxLength": 131072 + } + }, + { + "type": "array", + "minItems": 1, + "items": { + "$ref": "#/components/schemas/InputContentPart" + } + } + ] + } + } + }, + "InputContentPart": { + "type": "object", + "required": [ + "type", + "text" + ], + "properties": { + "type": { + "type": "string", + "const": "text", + "default": "text" + }, + "text": { + "type": "string", + "minLength": 1, + "format": "textarea", + "maxLength": 131072, + "description": "Text content of this part." + } + } + }, + "ModerationCategories": { + "type": "object", + "description": "Per-category boolean verdicts. All 13 categories are always present, in OpenAI's order.", + "properties": { + "harassment": { + "type": "boolean", + "description": "Content that expresses, incites, or promotes harassing language towards any target.", + "example": true + }, + "harassment/threatening": { + "type": "boolean", + "description": "Harassment that also includes violence or serious harm towards any target.", + "example": true + }, + "hate": { + "type": "boolean", + "description": "Content that expresses, incites, or promotes hate based on a protected attribute.", + "example": false + }, + "hate/threatening": { + "type": "boolean", + "description": "Hateful content that also includes violence or serious harm towards a protected group.", + "example": false + }, + "illicit": { + "type": "boolean", + "description": "Content that gives advice or instruction on how to commit a wrongdoing.", + "example": false + }, + "illicit/violent": { + "type": "boolean", + "description": "Illicit content that also references violence or procuring a weapon.", + "example": false + }, + "self-harm": { + "type": "boolean", + "description": "Content that promotes, encourages, or depicts acts of self-harm.", + "example": false + }, + "self-harm/intent": { + "type": "boolean", + "description": "Content where the speaker expresses that they are engaging or intend to engage in self-harm.", + "example": false + }, + "self-harm/instructions": { + "type": "boolean", + "description": "Content that provides instructions or advice on how to commit acts of self-harm.", + "example": false + }, + "sexual": { + "type": "boolean", + "description": "Content meant to arouse sexual excitement or that promotes sexual services.", + "example": false + }, + "sexual/minors": { + "type": "boolean", + "description": "Sexual content that includes an individual under 18 years old.", + "example": false + }, + "violence": { + "type": "boolean", + "description": "Content that depicts death, violence, or physical injury.", + "example": true + }, + "violence/graphic": { + "type": "boolean", + "description": "Content that depicts death, violence, or physical injury in graphic detail.", + "example": false + } + } + }, + "ModerationCategoryScores": { + "type": "object", + "description": "Per-category confidence scores in [0, 1]. All 13 categories are always present, in OpenAI's order.", + "properties": { + "harassment": { + "type": "number", + "description": "Confidence score for `harassment`.", + "example": 0.9412 + }, + "harassment/threatening": { + "type": "number", + "description": "Confidence score for `harassment/threatening`.", + "example": 0.9016 + }, + "hate": { + "type": "number", + "description": "Confidence score for `hate`.", + "example": 0.0231 + }, + "hate/threatening": { + "type": "number", + "description": "Confidence score for `hate/threatening`.", + "example": 0.0104 + }, + "illicit": { + "type": "number", + "description": "Confidence score for `illicit`.", + "example": 0.0057 + }, + "illicit/violent": { + "type": "number", + "description": "Confidence score for `illicit/violent`.", + "example": 0.0039 + }, + "self-harm": { + "type": "number", + "description": "Confidence score for `self-harm`.", + "example": 0.0021 + }, + "self-harm/intent": { + "type": "number", + "description": "Confidence score for `self-harm/intent`.", + "example": 0.0012 + }, + "self-harm/instructions": { + "type": "number", + "description": "Confidence score for `self-harm/instructions`.", + "example": 0.0008 + }, + "sexual": { + "type": "number", + "description": "Confidence score for `sexual`.", + "example": 0.0006 + }, + "sexual/minors": { + "type": "number", + "description": "Confidence score for `sexual/minors`.", + "example": 0.0002 + }, + "violence": { + "type": "number", + "description": "Confidence score for `violence`.", + "example": 0.8774 + }, + "violence/graphic": { + "type": "number", + "description": "Confidence score for `violence/graphic`.", + "example": 0.0311 + } + } + }, + "ModerationAppliedInputTypes": { + "type": "object", + "description": "Which input modality triggered each category. This is a text-only model, so every category maps to [\"text\"].", + "properties": { + "harassment": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "harassment/threatening": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "hate": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "hate/threatening": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "illicit": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "illicit/violent": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "self-harm": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "self-harm/intent": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "self-harm/instructions": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "sexual": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "sexual/minors": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "violence": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + }, + "violence/graphic": { + "type": "array", + "items": { + "type": "string" + }, + "example": [ + "text" + ] + } + } + }, + "ModerationResult": { + "type": "object", + "description": "A single entry of the `results` array — the verdict for one input.", + "properties": { + "flagged": { + "type": "boolean", + "description": "True when the model flagged the input in one or more categories.", + "example": true + }, + "categories": { + "$ref": "#/components/schemas/ModerationCategories" + }, + "category_scores": { + "$ref": "#/components/schemas/ModerationCategoryScores" + }, + "category_applied_input_types": { + "$ref": "#/components/schemas/ModerationAppliedInputTypes" + } + } + }, + "ModerationResponse": { + "type": "object", + "description": "OpenAI-compatible moderations envelope. One `results` entry per input.", + "properties": { + "id": { + "type": "string", + "description": "Unique identifier for the moderation request.", + "example": "modr-0a1b2c3d4e5f60718293a4b5c6d7e8f90" + }, + "model": { + "type": "string", + "description": "The model used for classification.", + "example": "zlm-v1-moderation-edge" + }, + "results": { + "type": "array", + "description": "Moderation verdicts, one per input (a string input yields a single-element array).", + "items": { + "$ref": "#/components/schemas/ModerationResult" + } + } + } + }, + "ErrorResponse": { + "type": "object", + "additionalProperties": true + }, + "SystemOneRequest": { + "type": "object", + "required": [ + "model", + "state", + "questions" + ], + "properties": { + "model": { + "type": "string", + "const": "d1.1", + "default": "d1.1", + "example": "d1.1", + "description": "Model identifier (fixed for this playground)." + }, + "state": { + "description": "The content every question refers to: a string, object or array. An object or array is read as its JSON. For `zlm-v1-iab-domain-classifier` it must be a URL or domain.", + "anyOf": [ + { + "type": "string" + }, + { + "type": "object" + }, + { + "type": "array", + "items": {} + } + ] + }, + "questions": { + "type": "object", + "minProperties": 1, + "maxProperties": 16, + "description": "Named questions (up to 16). Keys are names you choose; each answer comes back under the same key.", + "additionalProperties": { + "$ref": "#/components/schemas/SystemOneQuestion" + } + } + } + }, + "SystemOneQuestion": { + "oneOf": [ + { + "$ref": "#/components/schemas/NoulQuestion" + }, + { + "$ref": "#/components/schemas/ChoiceQuestion" + }, + { + "$ref": "#/components/schemas/ScoreQuestion" + } + ], + "discriminator": { + "propertyName": "type", + "mapping": { + "noul": "#/components/schemas/NoulQuestion", + "choice": "#/components/schemas/ChoiceQuestion", + "score": "#/components/schemas/ScoreQuestion" + } + } + }, + "NoulQuestion": { + "type": "object", + "required": [ + "type" + ], + "description": "A yes/no (true/false) decision.", + "properties": { + "type": { + "type": "string", + "const": "noul" + }, + "instructions": { + "description": "The statement or question to evaluate.", + "anyOf": [ + { + "type": "string" + }, + { + "type": "object" + }, + { + "type": "array", + "items": {} + }, + { + "type": "null" + } + ] + }, + "criteria": { + "description": "Optional definitions of what counts as true and false.", + "anyOf": [ + { + "type": "object", + "properties": { + "true": { + "description": "Free-form text: a string, or an object or array that is read as its JSON.", + "anyOf": [ + { + "type": "string" + }, + { + "type": "object" + }, + { + "type": "array", + "items": {} + }, + { + "type": "null" + } + ] + }, + "false": { + "description": "Free-form text: a string, or an object or array that is read as its JSON.", + "anyOf": [ + { + "type": "string" + }, + { + "type": "object" + }, + { + "type": "array", + "items": {} + }, + { + "type": "null" + } + ] + } + } + }, + { + "type": "null" + } + ] + } + } + }, + "ChoiceQuestion": { + "type": "object", + "required": [ + "type", + "criteria" + ], + "description": "Pick one option.", + "properties": { + "type": { + "type": "string", + "const": "choice" + }, + "instructions": { + "description": "Free-form text: a string, or an object or array that is read as its JSON.", + "anyOf": [ + { + "type": "string" + }, + { + "type": "object" + }, + { + "type": "array", + "items": {} + }, + { + "type": "null" + } + ] + }, + "criteria": { + "type": "object", + "minProperties": 1, + "maxProperties": 32, + "description": "Choice names mapped to descriptions. A null description means the name alone is used.", + "additionalProperties": { + "description": "Free-form text: a string, or an object or array that is read as its JSON.", + "anyOf": [ + { + "type": "string" + }, + { + "type": "object" + }, + { + "type": "array", + "items": {} + }, + { + "type": "null" + } + ] + } + } + } + }, + "ScoreQuestion": { + "type": "object", + "required": [ + "type", + "criteria" + ], + "description": "Rate against an ordered rubric.", + "properties": { + "type": { + "type": "string", + "const": "score" + }, + "instructions": { + "description": "What to rate.", + "anyOf": [ + { + "type": "string" + }, + { + "type": "object" + }, + { + "type": "array", + "items": {} + }, + { + "type": "null" + } + ] + }, + "criteria": { + "type": "array", + "minItems": 1, + "maxItems": 32, + "description": "Ordered level descriptions. A level's position is its score, starting at 0.", + "items": { + "anyOf": [ + { + "type": "string" + }, + { + "type": "object" + }, + { + "type": "array", + "items": {} + } + ] + } + } + } + }, + "SystemOneResponse": { + "type": "object", + "required": [ + "model", + "answers", + "usage" + ], + "properties": { + "model": { + "type": "string", + "description": "The model that answered." + }, + "answers": { + "type": "object", + "description": "Keyed by your question names, in request order.", + "additionalProperties": { + "$ref": "#/components/schemas/SystemOneAnswer" + } + }, + "usage": { + "$ref": "#/components/schemas/SystemOneUsage" + } + } + }, + "SystemOneAnswer": { + "oneOf": [ + { + "$ref": "#/components/schemas/NoulAnswer" + }, + { + "$ref": "#/components/schemas/ChoiceAnswer" + }, + { + "$ref": "#/components/schemas/ScoreAnswer" + } + ], + "discriminator": { + "propertyName": "type", + "mapping": { + "noul": "#/components/schemas/NoulAnswer", + "choice": "#/components/schemas/ChoiceAnswer", + "score": "#/components/schemas/ScoreAnswer" + } + } + }, + "NoulAnswer": { + "type": "object", + "required": [ + "type", + "noul" + ], + "properties": { + "type": { + "type": "string", + "const": "noul" + }, + "noul": { + "type": "number", + "minimum": 0, + "maximum": 1, + "description": "Probability of true. Near 1 favors true, near 0 favors false, near 0.5 is uncertain." + } + } + }, + "ChoiceAnswer": { + "type": "object", + "required": [ + "type", + "choice", + "confidence", + "probabilities" + ], + "properties": { + "type": { + "type": "string", + "const": "choice" + }, + "choice": { + "type": "string", + "description": "The most likely choice name." + }, + "confidence": { + "type": "number", + "minimum": 0, + "maximum": 1 + }, + "probabilities": { + "type": "object", + "description": "Keyed by choice name; sums to about 1.", + "additionalProperties": { + "type": "number" + } + } + } + }, + "ScoreAnswer": { + "type": "object", + "required": [ + "type", + "score", + "confidence", + "legend", + "probabilities" + ], + "properties": { + "type": { + "type": "string", + "const": "score" + }, + "score": { + "type": "number", + "description": "Probability-weighted average level; may fall between integers." + }, + "confidence": { + "type": "number", + "minimum": 0, + "maximum": 1 + }, + "legend": { + "type": "object", + "description": "Level key (\"0\", \"1\", ...) to its criteria text.", + "additionalProperties": {} + }, + "probabilities": { + "type": "object", + "description": "Keyed like `legend`; sums to about 1.", + "additionalProperties": { + "type": "number" + } + } + } + }, + "SystemOneUsage": { + "type": "object", + "required": [ + "input_tokens", + "output_tokens", + "total_tokens" + ], + "properties": { + "input_tokens": { + "type": "integer", + "description": "Tokens billed: the state plus the questions." + }, + "output_tokens": { + "type": "integer", + "description": "Always 0; System One output is not billed." + }, + "total_tokens": { + "type": "integer" + } + } + }, + "HTTPValidationError": { + "type": "object", + "description": "Request validation failure (System One's own error shape).", + "properties": { + "detail": { + "type": "array", + "items": { + "type": "object", + "required": [ + "loc", + "msg", + "type" + ], + "properties": { + "loc": { + "type": "array", + "items": { + "anyOf": [ + { + "type": "string" + }, + { + "type": "integer" + } + ] + } + }, + "msg": { + "type": "string" + }, + "type": { + "type": "string" + } + } + } + } + } + } + } + } +} diff --git a/api-reference/openapi/zerogpu.openapi.json b/api-reference/openapi/zerogpu.openapi.json index 61bd05c..eab299f 100644 --- a/api-reference/openapi/zerogpu.openapi.json +++ b/api-reference/openapi/zerogpu.openapi.json @@ -2321,7 +2321,7 @@ "properties": { "model": { "type": "string", - "description": "A model that supports System One: `zlm-v1-moderation-edge`, `zlm-v1-iab-classify-edge`, `zlm-v1-iab-domain-classifier` or `zlm-v2-iab-classify-edge-enriched`.", + "description": "A model that supports System One: `d1.1`, `zlm-v1-moderation-edge`, `zlm-v1-iab-classify-edge`, `zlm-v1-iab-domain-classifier` or `zlm-v2-iab-classify-edge-enriched`.", "example": "zlm-v1-iab-classify-edge" }, "state": { diff --git a/api-reference/systemone.mdx b/api-reference/systemone.mdx index 1a0996c..94143fc 100644 --- a/api-reference/systemone.mdx +++ b/api-reference/systemone.mdx @@ -7,10 +7,11 @@ openapi: "api-reference/openapi/zerogpu.openapi.json POST /systemone" Send the content as `state` and ask up to 16 named `questions` about it in one call. Each answer comes back under the question's name, with calibrated probabilities. The request and response follow the [TypeSafe System One](https://api.typesafe.ai/openapi.json) decision schema, so an existing System One client works by changing the base URL and API key. -System One is available on four models, in addition to their existing endpoints: +System One is available on five models. For `d1.1` it is the only endpoint; the ZLM classifiers also keep their existing ones: | Model | Also served on | Good for | | --- | --- | --- | +| [`d1.1`](/api-reference/models/d1-1) | System One only | Does this need a human? Which intent is it? How urgent is it? | | [`zlm-v1-moderation-edge`](/api-reference/models/zlm-v1-moderation-edge) | `/moderations` | Is this unsafe? Which moderation category applies? How severe is it? | | [`zlm-v1-iab-classify-edge`](/api-reference/models/zlm-v1-iab-classify-edge) | `/chat/completions`, `/responses` | Which IAB category matches? Does it belong to category X? How strongly? | | [`zlm-v1-iab-domain-classifier`](/api-reference/models/zlm-v1-iab-domain-classifier) | `/chat/completions`, `/responses` | The same questions about a domain. `state` must be a URL or domain. | @@ -28,6 +29,8 @@ Each model scores the content against its own labels: the 13 OpenAI moderation c A question whose criteria are not among the model's labels is answered by zero-shot classification instead. You can mix both kinds in one request. +`d1.1` has no fixed label set of its own. It is a decision model, so every question is answered in a single forward pass from the `state` and the question's own criteria. + ## Limits and billing - Up to 16 questions per request, and up to 32 criteria per question. diff --git a/docs.json b/docs.json index 9b3b30c..20caa41 100644 --- a/docs.json +++ b/docs.json @@ -113,6 +113,7 @@ "api-reference/models/gpt-5-6-luna", "api-reference/models/gpt-6-luna", "api-reference/models/gpt-4-1-mini", + "api-reference/models/d1-1", "api-reference/models/deberta-v3-small", "api-reference/models/gliner2-5-base-v1", "api-reference/models/gliner2-5-multi-v1", diff --git a/docs/model-catalog.mdx b/docs/model-catalog.mdx index 2a69216..cf1b647 100644 --- a/docs/model-catalog.mdx +++ b/docs/model-catalog.mdx @@ -22,6 +22,7 @@ Sections on this page: [At a glance](#at-a-glance) (pricing table), [Detailed mo | llama-3.1-8b-instruct-fast llama-3.1-8b-instruct-fast | $0.15 | $0.28 | $0.025 | Text Generation | 131,072 | | llama-3.2-3b-instruct-fast llama-3.2-3b-instruct-fast | $0.05 | $0.30 | $0.025 | Text Generation | 131,072 | | llama-guard-4-12b llama-guard-4-12b | $0.18 | $0.18 | — | Text Generation | 163,840 | +| d1.1 d1.1 | $0.015 | Not billed | — | Text Classification | 65,536 | | zlm-v2-iab-classify-edge-enriched zlm-v2-iab-classify-edge-enriched | $0.05 | $0.15 | — | Text Classification | 800 | | zlm-v1-iab-classify-edge zlm-v1-iab-classify-edge | $0.02 | $0.05 | — | Text Classification | 400 | | zlm-v1-iab-domain-classifier zlm-v1-iab-domain-classifier | $0.02 | $0.05 | — | Text Classification | 100 | @@ -35,7 +36,7 @@ Sections on this page: [At a glance](#at-a-glance) (pricing table), [Detailed mo | zlm-v1-tool-decisions | $0.05 | $0.15 | — | Text Generation | 24,576 | | unlimited-ocr unlimited-ocr | Per page | Per page | — | Text Generation | — | | zlm-v1-signal-extract zlm-v1-signal-extract | $0.04 | $0.10 | — | Text Classification | 400 | -| gliner2.5-decide-v1 gliner2.5-decide-v1 | $0.004 | Not billed | — | Text Classification | — | +| gliner2.5-decide-v1 gliner2.5-decide-v1 | $0.01 | Not billed | — | Text Classification | — | | all-minilm-l6-v2 all-minilm-l6-v2 | $0.004 | Not billed | — | Text Embedding | 512 | | bge-small-en-v1.5 bge-small-en-v1.5 | $0.004 | Not billed | — | Text Embedding | 512 | @@ -162,6 +163,16 @@ Sections on this page: [At a glance](#at-a-glance) (pricing table), [Detailed mo Meta's Llama Guard 4 12B is a multimodal safety classification model for moderating text, images, and mixed text-image inputs. It evaluates both incoming prompts… + +
+ d1.1 +
d1.1
+
+ +
Text ClassificationDecisionsSystem One65,536 context window\$0.015 / 1M input
+ + LiquidAI's d1.1 is a decision model and the successor to d1. Send a state — text, JSON, or images — together with several named questions in one System One call, and every question is answered in a single forward pass: noul returns the probability of yes, choice picks one of your named options… +
zlm-v2-iab-classify-edge-enriched @@ -287,7 +298,7 @@ Sections on this page: [At a glance](#at-a-glance) (pricing table), [Detailed mo
gliner2.5-decide-v1
-
486M params\$0.004 / 1M input
+
486M params\$0.01 / 1M input
Fastino's GLiNER2.5 Decide is a 486M-parameter zero-shot decision and classification model built for operational AI workflows. You supply the labels or decision options at request time to classify intent, route agents and tools, triage support requests, make moderation decisions…
@@ -366,7 +377,7 @@ Sections on this page: [At a glance](#at-a-glance) (pricing table), [Detailed mo 3 models: `whisper-tiny`, `chatterbox-nano`, `chatterbox` - 8 models: `zlm-v2-iab-classify-edge-enriched`, `zlm-v1-iab-classify-edge`, `zlm-v1-iab-domain-classifier`, `deberta-v3-small`, `zlm-v1-signal-extract`, `gliner2.5-base-v1`, `gliner2.5-multi-v1`, `gliner2.5-decide-v1` + 9 models: `d1.1`, `zlm-v2-iab-classify-edge-enriched`, `zlm-v1-iab-classify-edge`, `zlm-v1-iab-domain-classifier`, `deberta-v3-small`, `zlm-v1-signal-extract`, `gliner2.5-base-v1`, `gliner2.5-multi-v1`, `gliner2.5-decide-v1` 1 model: `zlm-v1-moderation-edge` diff --git a/docs/text-classification.mdx b/docs/text-classification.mdx index 89790d1..8fc4bc4 100644 --- a/docs/text-classification.mdx +++ b/docs/text-classification.mdx @@ -3,10 +3,11 @@ title: "Text Classification" description: "Edge models that map text to labels, IAB taxonomy, and audience signals." --- -Text classification models map a passage to predefined labels or categories. ZeroGPU offers IAB taxonomy classifiers tuned for ad tech, a general zero-shot classifier where you supply the candidate labels, and the schema-based GLiNER2.5 models, which classify, extract entities, fill a JSON schema, and make dynamic-label decisions such as intent and routing from the same request. Each request is shown for both the [Responses API](/api-reference/responses) and the OpenAI-compatible [Chat Completions API](/api-reference/chat-completions). +Text classification models map a passage to predefined labels or categories. ZeroGPU offers IAB taxonomy classifiers tuned for ad tech, a general zero-shot classifier where you supply the candidate labels, and the schema-based GLiNER2.5 models, which classify, extract entities, fill a JSON schema, and make dynamic-label decisions such as intent and routing from the same request. LiquidAI's `d1.1` decision model sits alongside them, answering named yes/no, multiple-choice, and rubric questions about any state. Each request is shown for both the [Responses API](/api-reference/responses) and the OpenAI-compatible [Chat Completions API](/api-reference/chat-completions); `d1.1` is served only on [System One](/api-reference/systemone). | Model | Input /1M | Output /1M | Max tokens | | --- | ---: | ---: | ---: | +| d1.1 d1.1 | $0.015 | Not billed | 65,536 | | zlm-v1-iab-classify-edge zlm-v1-iab-classify-edge | $0.02 | $0.05 | 400 | | zlm-v2-iab-classify-edge-enriched zlm-v2-iab-classify-edge-enriched | $0.05 | $0.15 | 800 | | zlm-v1-iab-domain-classifier zlm-v1-iab-domain-classifier | $0.02 | $0.05 | 100 | @@ -14,7 +15,57 @@ Text classification models map a passage to predefined labels or categories. Zer | zlm-v1-signal-extract zlm-v1-signal-extract | $0.04 | $0.10 | 400 | | gliner2.5-base-v1 gliner2.5-base-v1 | $0.02 | $0.05 | — | | gliner2.5-multi-v1 gliner2.5-multi-v1 | $0.02 | $0.05 | — | -| gliner2.5-decide-v1 gliner2.5-decide-v1 | $0.004 | Not billed | — | +| gliner2.5-decide-v1 gliner2.5-decide-v1 | $0.01 | Not billed | — | + +## d1.1 + + + This model is routable only on `/v1/systemone` (see + [System One](/api-reference/systemone)). It does not chat, stream, or call + tools, so it 400s on `/v1/responses` and `/v1/chat/completions`. Send the + content to judge as `state` and up to 16 named `questions` about it. + + +> LiquidAI's d1.1 is a decision model and the successor to d1. Send a state — text, JSON, or images — together with several named questions, and every question is answered in a single forward pass: `noul` returns the probability of yes, `choice` picks one of your named options with a confidence and the full probability distribution, and `score` rates the state on 2 to 10 ordered levels. It generates no text, so output tokens are always 0 and only the input is billed, across a 65,536-token state. It is built for intent classification, agent routing, moderation, commercial-intent detection, triage, and LLM-as-a-judge checks, and it does not chat, stream, or call tools. + +**References:** [Terms](https://www.liquid.ai/terms-conditions) • [Privacy](https://zerogpu.ai/privacy-policy) + + + +```bash System One +curl https://api.zerogpu.ai/v1/systemone \ + -H "content-type: application/json" \ + -H "x-api-key: YOUR_API_KEY" \ + -d '{ + "model": "d1.1", + "state": "Can you refund my last order? It arrived broken and I want to speak to someone.", + "questions": { + "handoff": { + "type": "noul", + "instructions": "Does this request need a human agent?" + }, + "intent": { + "type": "choice", + "instructions": "Which intent best describes this request?", + "criteria": { + "refund": "Asking for money back", + "payment_issue": "A payment failed or was charged twice", + "account_access": "Cannot sign in or recover an account", + "general_question": "Anything else" + } + }, + "urgency": { + "type": "score", + "instructions": "How urgent is this request?", + "criteria": ["None", "Low", "Medium", "High"] + } + } +}' +``` + + + +Each answer comes back under its question's name, with calibrated probabilities and a `usage` block whose `output_tokens` is always 0 — see [System One](/api-reference/systemone) for the answer shapes of `noul`, `choice`, and `score`. ## zlm-v1-iab-classify-edge