venice-chat
Call POST /chat/completions on Venice. Covers the OpenAI-compatible request shape, Venice-only venice_parameters (web search, E2EE, characters, thinking control, X search), multimodal inputs (images/audio/video), tool calls, reasoning controls, streaming, prompt caching, structured output, and model feature suffixes.
veniceai · 已有 AI 审核记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add veniceai/skills --skill venice-chat技能说明
name: venice-chat description: Call POST /chat/completions on Venice. Covers the OpenAI-compatible request shape, Venice-only venice_parameters (web search, E2EE, characters, thinking control, X search), multimodal inputs (images/audio/video), tool calls, reasoning controls, streaming, prompt caching, structured output, and model feature suffixes.
Venice Chat Completions
POST /api/v1/chat/completions is Venice's main text endpoint. It's OpenAI-compatible, plus a venice_parameters object for Venice-only features.
Use when
- You need LLM text generation, with or without tools, with or without streaming.
- You want multimodal inputs (images, audio, video) to a vision/audio-capable model.
- You want Venice-specific features: web search, E2EE, characters, xAI X/Twitter search, strip-thinking, web scraping.
- You need prompt caching for large system prompts or long documents.
- You need structured (
json_schema) output.
For the newer Alpha Responses API, see venice-responses.
Minimal request
curl https://api.venice.ai/api/v1/chat/completions \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "zai-org-glm-5-1",
"messages": [{"role": "user", "content": "Why is the sky blue?"}]
}'
Response shape is the standard OpenAI chat.completion object (id, object: "chat.completion", choices[].message, usage). With stream: true, responses come as SSE data: lines in chat.completion.chunk format.
The request body
Core fields (OpenAI-compatible)
| Field | Notes |
|---|---|
model | string — model ID, trait name, or compatibility mapping. Suffixes allowed (see below). Required. |
messages | array of system / developer / user / assistant / tool messages. Required, min 1. |
temperature, top_p, top_k, min_p, min_temp, max_temp | sampling controls |
repetition_penalty, frequency_penalty, presence_penalty | repetition controls |
max_tokens (deprecated) / max_completion_tokens | upper bound on output tokens |
n | number of choices (keep 1 to minimize cost) |
seed | integer for reproducibility |
stop / stop_token_ids | up to 4 strings, or raw token IDs |
stream, stream_options.include_usage | SSE streaming + include usage in the final chunk |
response_format | {type:"json_schema", json_schema:{...}} (preferred), {type:"json_object"}, or {type:"text"} |
tools, tool_choice, parallel_tool_calls | function calling / built-in tools |
logprobs, top_logprobs | return token log-probabilities |
reasoning.effort / reasoning_effort | none | minimal | low | medium | high | xhigh | max |
reasoning.summary | auto | concise | detailed |
prompt_cache_key, prompt_cache_retention (default/extended/24h) | prompt caching hints. extended and 24h both extend retention to 24 hours on supported models |
verbosity, text.verbosity | low/medium/high/auto. Also accepted as a root-level field, not only nested under text |
include | array of extra fields to include in the response (OpenAI compat) |
fallbacks | up to 10 entries. Anthropic beta parameter for Claude Fable 5 server-side refusal fallback. Forwarded only on direct Anthropic routes and ignored by every other provider |
metadata | key/value strings for tracking |
user, store | accepted but ignored (OpenAI compat) |
venice_parameters (Venice-only)
All optional. Combined with model feature suffixes, these are how you enable Venice features.
| Field | Type | Default | Effect |
|---|---|---|---|
character_slug | string | — | Apply a published Venice character. Slug is the "Public ID" on the character page. See venice-characters. |
strip_thinking_response | bool | false | Strip <think>...</think> from the assistant output on reasoning models. |
disable_thinking | bool | false | Disable thinking entirely on supported reasoning models and strip tags. |
enable_e2ee | bool | true | End-to-end encryption on E2EE-capable models when E2EE headers are present. Set to false to force TEE-only. |
enable_web_search | "off"/"auto"/"on" | "off" | Venice server-side web search. Citations arrive in the first streamed chunk or the response. |
enable_web_scraping | bool | false | Scrape any URLs found in the last user message (Firecrawl). |
enable_web_citations | bool | false | Ask the LLM to cite sources with ^1^ / ^1,3^ superscripts. |
include_search_results_in_stream | bool | false | Experimental — emit search results as the first stream chunk. |
return_search_results_as_documents | bool | — | Also surface search results as a synthetic tool call venice_web_search_documents (LangChain-friendly). |
include_venice_system_prompt | bool | true | Prepend Venice's curated system prompt. Turn off for full control. |
enable_x_search | bool | false | xAI native web + X/Twitter search (Grok models with supportsXSearch). Adds ~$0.01/search. |
Model feature suffixes
Some venice_parameters can also be expressed as model feature suffixes on the model string — useful when the caller/library (OpenAI SDK, LangChain) can't set venice_parameters. Syntax:
<model-id>:<key>=<value>[&<key>=<value>…]
Values are URL-decoded. Supported keys (exact match):
| Key | Type | Maps to |
|---|---|---|
enable_web_search | on / off / auto | venice_parameters.enable_web_search |
enable_web_citations | "true" / "false" | venice_parameters.enable_web_citations |
enable_web_scraping | "true" / "false" | venice_parameters.enable_web_scraping |
include_venice_system_prompt | "true" / "false" | venice_parameters.include_venice_system_prompt |
include_search_results_in_stream | "true" / "false" | `venice_paramet |