Chat Completions

Our Agent (chat completions) endpoint is designed to be compatible with the OpenAI chat completions API specification, so most OpenAI SDKs and tools work with minimal changes. Agent also extends the surface with custom response_format values, conversation metadata, Bible/language controls, and declaration-only tools. It does not currently support multi-modal input or output, and returns a single completion at a time. See the OpenAI SDK libraries for client options.

Demo Chatbot App

Minimal Request Example #

Since your Agent is configured on the platform, API calls to get completions can be extremely minimal. All that’s required is the prompt for which you want a response:

curl \
--header 'x-api-key: apg_xxxxxxxxxxxxxxxxxxxxxxxxxxxx' \
--header 'Content-Type: application/json' \
--data '{
    "prompt": "How can a good God allow so much evil in the world?"
}' \
--url https://my.gospel.bot/api/v1/chat/completions

Please note: you must replace the x-api-key value with your API key and my.gospel.bot with your Agent’s domain.

Full Request Example #

However, the API supports overriding the default Agent configuration options, as well as several other runtime options. Here’s an example of all supported parameters that have an effect on the output:

curl \
--header 'Authorization: Bearer apg_xxxxxxxxxxxxxxxxxxxxxxxxxxxx' \
--header 'Content-Type: application/json' \
--data '{
    "model": "openai/gpt/4o",
    "stream": false,
    "messages": [
        {
            "role": "system",
            "content": "This a system prompt override."
        },
        {
            "role": "user",
            "content": "This is a previous prompt."
        },
        {
            "role": "assistant",
            "content": "This is a previous completion."
        },
        {
            "role": "user",
            "content": "This is the current prompt."
        }
    ],
    "response_format": { 
        "type": "json" 
    },
    "metadata": {
        "anonymous": true,
        "conversation": null,
        "language": "en",
        "session": null,
        "device": null,
        "bible": "esv",
        "variables": {
            "name": "Jake"
        }
    },
    "frequency_penalty": 0.25,
    "presence_penalty": -0.25,
    "max_completion_tokens": 1024,
    "reasoning_effort": "high",
    "verbosity": "medium",
    "temperature": 0.5,
    "top_p": 0.9,
    "user": null
}' \
--url https://my.gospel.bot/api/v1/chat/completions

Please note: you must replace the Authorization bearer token value with your API key and my.gospel.bot with your Agent’s domain.

Authentication #

See Authentication on the Overview page for how to authenticate Fusion API requests.

Same-origin browser requests to the Agent host may omit an API key when creating a completion (POST). The list and get endpoints (GET /api/v1/chat/completions and GET /api/v1/chat/completions/{id}) always require a valid API key.

Prompt and Messages #

You must either supply the prompt string or a messages array.

The messages parameter is an array of message objects, each with a role and a content property. By default, an Agent will have a system prompt that is automatically applied. Also by default, previous exchanges between an Agent and user as defined by the metadata.device, metadata.session, metadata.conversation, and user parameters are prepended to the array of messages which are eventually sent to the LLM. You may control how many past exchanges you wish Agent to use by passing an integer for the metadata.max_memories parameter. You only need supply a messages array if you wish to override this default behavior.

If you want to use the aforementioned defaults, or the response doesn’t require context from previous exchanges with Agent, you may simply provide a prompt string.

If you wish to prevent any context of previous exchanges altogether, regardless of the value of metadata.device, metadata.session, metadata.conversation, and user — you may pass metadata.anonymous as true.

• user: any string identifier for the user (100 characters max)
• metadata.conversation: any string identifier for the conversation within a single session (100 characters max)
• metadata.session: any string identifier for the user’s session (100 characters max)
• metadata.device: any string identifier for the user’s device (100 characters max)
• metadata.anonymous: if set to true, no past exchange context is provided to the Agent (one-shot); equivalent to setting metadata.max_memories: 0
• metadata.max_memories: the number of previous exchanges to provide to Agent
• metadata.variables: object of string key/value pairs substituted into {key} placeholders in the assembled system prompt (max 50 keys, 500 characters per value; not persisted). Reserved keys cannot be overridden: language, bible, translation, passages, year, month, day, date, time, country, region, city, timezone

Streaming vs Non-Streaming #

The stream parameter controls whether the response should be streamed or delivered all at once. Typically, chat interfaces benefit from streaming output as the user gets more instantaneous feedback. However, there are use cases that lend themselves to non-streaming output as well.

Response Format #

The response format may be specified using the response_format.type parameter. Valid values are as follows:

• raw: raw OpenAI compatible chunked chat completion JSON. Only available if the stream option is true. Setting this format will allow any OpenAI compatible library / SDK to be used against this endpoint, as long as it has a way to set the response_format parameter. [default when streaming]
  
• text: plain text output; may include markdown
  
• html: HTML formatted, converted from markdown where applicable
  
• json: structured JSON response; includes completion as well as token usage and timing stats. Only available if the stream option is false. [default when not streaming]
• json_schema: structured output constrained to a supplied JSON Schema under response_format.json_schema.schema (optional name, description, strict). API-only and non-streaming (stream: true is ignored). Missing schema returns 422. Bypasses the prose completion cache.

When streaming with raw, the initial chunk may include a top-level metadata object (sources, conversation/share fields, and when present guardrail / cta). The final chunk reports token usage, and may include url_citation annotations on the delta.

Structured Outputs (json_schema) #

Example request:

{
    "messages": [{"role": "user", "content": "Give me a person."}],
    "response_format": {
        "type": "json_schema",
        "json_schema": {
            "name": "person",
            "schema": {
                "type": "object",
                "properties": {
                    "name": {"type": "string"},
                    "age": {"type": "integer"}
                },
                "required": ["name", "age"]
            }
        }
    }
}

Your Agent can be configured via Apologist Ignite to use any of the above response formats above by default for the chat completions endpoint. This is especially useful in situations where a 3rd party integration doesn’t allow sending custom parameters. Note that if text is specified, you may also toggle an option in your Agent configuration in Apologist Ignite to automatically strip all markdown to ensure only plain text is returned.

Bible Translation #

The metadata.bible parameter indicates which Bible translation the Agent should prefer. Pass a translation id or external_id string (up to 50 characters). Available translations are configured per Agent and provider (and may vary by language); they are not a fixed global enum. Common English examples include:

Language #

Use the metadata.language parameter to control the output language. Language support varies by model and is specific to a given Agent. However, any Agent can be upgraded to utilize real-time translation, expanding its language support to 192 languages. Here is the full list of supported languages:

Model #

Agent supports dozens of models across many providers using a standardized naming scheme: creator/family/variant. Pass the model name via the model parameter.

Limited Models #

Low-latency, inexpensive models with very limited reasoning abilities.

Model	Model ID	Supported Languages
OpenAI GPT-5.4 nano	openai/gpt/5.4-nano	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
xAI Grok 4 Fast	xai/grok/4-fast	ar, bn, cs, de, en, es, fa, fr, he, hi, id, it, ja, km, ko, lo, ms, my, nl, pl, pt, ru, th, tl, tr, ur, vi, zh
DeepSeek v3.1 Terminus	deepseek/deepseek/v3.1	af, am, ar, az, bg, bn, ca, cs, cy, da, de, el, en, eo, es, eu, fi, fr, fr-CA, ga, gd, gl, gu, ha, he, hi, hr, hu, hy, id, ig, is, it, ja, ka, kk, km, kn, ko, lo, ml, mn, mr, ms, mt, my, ne, nl, no, pa, pl, ps, pt, pt-BR, ro, ru, si, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, uz, vi, yo, zh, zh-TW, zu
OpenAI GPT-5.4 mini	openai/gpt/5.4-mini	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
Google Gemini 3 Flash	google/gemini/3-flash	ar, bg, bn, cs, da, de, el, en, es, et, fi, fr, he, hi, hr, hu, id, it, ja, ko, lt, lv, nl, no, pl, pt, ro, ru, sk, sl, sr, sv, sw, th, tr, uk, vi, zh, zh-TW
OpenAI GPT-5.6 Luna	openai/gpt/5.6-luna	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
OpenAI GPT OSS 120b	openai/gpt/oss-120b	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
Meta Llama 4 Scout	meta/llama4/scout	ar, de, en, es, fr, hi, id, it, ja, nl, pt, ru, th, tr, zh
Mistral Medium 3	mistral/medium/3	de, en, es, fr, it, ja, ko, nl, po, pt, zh
MiniMax M3	minimaxai/m3	af, ar, bg, ca, cs, da, de, el, en, es, fa, fi, fr, fr-CA, he, hi, hr, hu, id, it, ja, ko, ms, nl, nno, no, pl, pt, pt-BR, ro, ru, sk, sl, sv, ta, th, tl, tr, uk, vi, yue, zh, zh-TW
NVIDIA Nemotron 3 Ultra	nvidia/nemotron3/ultra	af, ar, as, awa, az, ba, ban, be, bg, bho, bn, bs, ca, ceb, cs, cy, da, de, el, en, es, et, eu, fa, fi, fr, fr-CA, ga, gl, gu, he, hi, hr, ht, hu, hy, id, ilo, is, it, ja, jw, ka, kk, km, kn, ko, lb, li, lij, lmo, lo, lt, lv, mai, min, mk, ml, mr, ms, mt, my, ne, nl, no, oc, or, pa, pag, pap, pl, pt, pt-BR, ro, ru, scn, sd, si, sk, sl, sq, sr, su, sv, sw, szl, ta, te, tg, th, tl, tr, tt, uk, ur, uz, vi, yi, yue, zh, zh-TW
Alibaba Qwen 3.7 Plus	alibaba/qwen3.7/plus	af, ar, as, awa, az, ba, ban, be, bg, bho, bn, bs, ca, ceb, cs, cy, da, de, el, en, es, et, eu, fa, fi, fr, fr-CA, ga, gl, gu, he, hi, hr, ht, hu, hy, id, ilo, is, it, ja, jw, ka, kk, km, kn, ko, lb, li, lij, lmo, lo, lt, lv, mai, min, mk, ml, mr, ms, mt, my, ne, nl, no, oc, or, pa, pag, pap, pl, pt, pt-BR, ro, ru, scn, sd, si, sk, sl, sq, sr, su, sv, sw, szl, ta, te, tg, th, tl, tr, tt, uk, ur, uz, vi, yi, yue, zh, zh-TW
Google Gemma 4 31B	google/gemma/gemma-4-31b	ar, bg, bn, cs, da, de, el, en, es, et, fi, fr, he, hi, hr, hu, id, it, ja, ko, lt, lv, nl, no, pl, pt, ro, ru, sk, sl, sr, sv, sw, th, tr, uk, vi, zh, zh-TW
Sarvam 30B	sarvam/sarvam/30b	as, bn, brx, doi, en, gu, hi, kn, kok, ks, mai, ml, mni, mr, ne, or, pa, sa, sat, sd, ta, te, ur
Alibaba Qwen 3.7 Max	alibaba/qwen3.7/max	af, ar, as, awa, az, ba, ban, be, bg, bho, bn, bs, ca, ceb, cs, cy, da, de, el, en, es, et, eu, fa, fi, fr, fr-CA, ga, gl, gu, he, hi, hr, ht, hu, hy, id, ilo, is, it, ja, jw, ka, kk, km, kn, ko, lb, li, lij, lmo, lo, lt, lv, mai, min, mk, ml, mr, ms, mt, my, ne, nl, no, oc, or, pa, pag, pap, pl, pt, pt-BR, ro, ru, scn, sd, si, sk, sl, sq, sr, su, sv, sw, szl, ta, te, tg, th, tl, tr, tt, uk, ur, uz, vi, yi, yue, zh, zh-TW
Meta Llama 4 Maverick	meta/llama4/maverick	ar, de, en, es, fr, hi, id, it, ja, nl, pt, ru, th, tr, zh
OpenAI GPT-4o mini	openai/gpt/4o-mini	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
OpenAI GPT-5 mini	openai/gpt/5-mini	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
Alibaba Qwen 3 Next 80B Instruct	alibaba/qwen3/next-80b-instruct	af, ar, as, awa, az, ba, ban, be, bg, bho, bn, bs, ca, ceb, cs, cy, da, de, el, en, es, et, eu, fa, fi, fr, fr-CA, ga, gl, gu, he, hi, hr, ht, hu, hy, id, ilo, is, it, ja, jw, ka, kk, km, kn, ko, lb, li, lij, lmo, lo, lt, lv, mai, min, mk, ml, mr, ms, mt, my, ne, nl, no, oc, or, pa, pag, pap, pl, pt, pt-BR, ro, ru, scn, sd, si, sk, sl, sq, sr, su, sv, sw, szl, ta, te, tg, th, tl, tr, tt, uk, ur, uz, vi, yi, yue, zh, zh-TW
Alibaba Qwen 3 235B	alibaba/qwen3/235b	af, ar, as, awa, az, ba, ban, be, bg, bho, bn, bs, ca, ceb, cs, cy, da, de, el, en, es, et, eu, fa, fi, fr, fr-CA, ga, gl, gu, he, hi, hr, ht, hu, hy, id, ilo, is, it, ja, jw, ka, kk, km, kn, ko, lb, li, lij, lmo, lo, lt, lv, mai, min, mk, ml, mr, ms, mt, my, ne, nl, no, oc, or, pa, pag, pap, pl, pt, pt-BR, ro, ru, scn, sd, si, sk, sl, sq, sr, su, sv, sw, szl, ta, te, tg, th, tl, tr, tt, uk, ur, uz, vi, yi, yue, zh, zh-TW
DeepSeek v3	deepseek/deepseek/v3	af, am, ar, az, bg, bn, ca, cs, cy, da, de, el, en, eo, es, eu, fi, fr, fr-CA, ga, gd, gl, gu, ha, he, hi, hr, hu, hy, id, ig, is, it, ja, ka, kk, km, kn, ko, lo, ml, mn, mr, ms, mt, my, ne, nl, no, pa, pl, ps, pt, pt-BR, ro, ru, si, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, uz, vi, yo, zh, zh-TW, zu
Alibaba Qwen3.5 397B A17b	alibaba/qwen3.5/397b-a17b	af, ar, as, awa, az, ba, ban, be, bg, bho, bn, bs, ca, ceb, cs, cy, da, de, el, en, es, et, eu, fa, fi, fr, fr-CA, ga, gl, gu, he, hi, hr, ht, hu, hy, id, ilo, is, it, ja, jw, ka, kk, km, kn, ko, lb, li, lij, lmo, lo, lt, lv, mai, min, mk, ml, mr, ms, mt, my, ne, nl, no, oc, or, pa, pag, pap, pl, pt, pt-BR, ro, ru, scn, sd, si, sk, sl, sq, sr, su, sv, sw, szl, ta, te, tg, th, tl, tr, tt, uk, ur, uz, vi, yi, yue, zh, zh-TW

Standard Models #

Economic models that balance lower cost with better responses.

Model	Model ID	Supported Languages
Anthropic Claude Haiku 4.5	anthropic/claude/haiku-4.5	ar, de, en, es, fr, hi, it, ja, ko, pt, ru, zh
xAI Grok 4.5	xai/grok/4.5	ar, bn, cs, de, en, es, fa, fr, he, hi, id, it, ja, km, ko, lo, ms, my, nl, pl, pt, ru, th, tl, tr, ur, vi, zh
DeepSeek V4 Pro	deepseek/deepseek/v4-pro	af, am, ar, az, bg, bn, ca, cs, cy, da, de, el, en, eo, es, eu, fi, fr, fr-CA, ga, gd, gl, gu, ha, he, hi, hr, hu, hy, id, ig, is, it, ja, ka, kk, km, kn, ko, lo, ml, mn, mr, ms, mt, my, ne, nl, no, pa, pl, ps, pt, pt-BR, ro, ru, si, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, uz, vi, yo, zh, zh-TW, zu
Google Gemini 3.5 Flash	google/gemini/3.5-flash	ar, bg, bn, cs, da, de, el, en, es, et, fi, fr, he, hi, hr, hu, id, it, ja, ko, lt, lv, nl, no, pl, pt, ro, ru, sk, sl, sr, sv, sw, th, tr, uk, vi, zh, zh-TW
Anthropic Claude Sonnet 5	anthropic/claude/sonnet-5	ar, bn, de, en, es, fr, hi, id, it, ja, ko, pt, ru, sw, yo, zh
OpenAI GPT-5.6 Terra	openai/gpt/5.6-terra	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
OpenAI o4 mini	openai/gpt/o4-mini	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
GLM 5.2	zai/glm/5.2	en, es, fr, ja, ko, ru, zh
Kimi K2.6 FP4	moonshotai/kimi/k2.6	af, am, ar, bg, bn, ceb, cs, da, de, el, en, es, et, fa, fi, fr, ga, gu, ha, he, hi, hr, hu, id, ig, is, it, ja, kk, km, kn, ko, lo, lt, lv, ml, mr, ms, mt, my, nb, ne, nl, pa, pa-Arab, pl, ps, pt, ro, ru, si, sk, so, sr, sv, sw, ta, te, th, tl, tr, uk, ur, uz, vi, xh, yo, zh, zh-TW, zu
Sarvam 105B	sarvam/sarvam/105b	as, bn, brx, doi, en, gu, hi, kn, kok, ks, mai, ml, mni, mr, ne, or, pa, sa, sat, sd, ta, te, ur
OpenAI GPT-5.1 Chat	openai/gpt/5.1	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
DeepSeek R1	deepseek/deepseek/r1	af, am, ar, az, bg, bn, ca, cs, cy, da, de, el, en, eo, es, eu, fi, fr, fr-CA, ga, gd, gl, gu, ha, he, hi, hr, hu, hy, id, ig, is, it, ja, ka, kk, km, kn, ko, lo, ml, mn, mr, ms, mt, my, ne, nl, no, pa, pl, ps, pt, pt-BR, ro, ru, si, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, uz, vi, yo, zh, zh-TW, zu
Google Gemini 3.5 Pro	google/gemini/3.5-pro	ar, bg, bn, cs, da, de, el, en, es, et, fi, fr, he, hi, hr, hu, id, it, ja, ko, lt, lv, nl, no, pl, pt, ro, ru, sk, sl, sr, sv, sw, th, tr, uk, vi, zh, zh-TW
OpenAI GPT-4.1	openai/gpt/4.1	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
OpenAI GPT-5 Chat	openai/gpt/5-chat	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
OpenAI o3	openai/gpt/o3	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
Google Gemini 3 Pro	google/gemini/3-pro	ar, bg, bn, cs, da, de, el, en, es, et, fi, fr, he, hi, hr, hu, id, it, ja, ko, lt, lv, nl, no, pl, pt, ro, ru, sk, sl, sr, sv, sw, th, tr, uk, vi, zh, zh-TW

Premium Models #

Higher quality output at a premium price point.

Model	Model ID	Supported Languages
OpenAI GPT-5.4	openai/gpt/5.4	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
Anthropic Claude Opus 5	anthropic/claude/opus-5	ar, bn, de, en, es, fr, hi, id, it, ja, ko, pt, ru, sw, yo, zh
OpenAI GPT-5.6 Sol	openai/gpt/5.6-sol	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
OpenAI GPT-5.3 Chat	openai/gpt/5.3	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
OpenAI GPT-5.5 Instant	openai/gpt/chat	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
OpenAI GPT-5.2 Chat	openai/gpt/5.2	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW
Anthropic Claude Sonnet 4.6	anthropic/claude/sonnet-4.6	ar, bn, de, en, es, fr, hi, id, it, ja, ko, pt, ru, sw, yo, zh
Anthropic Claude Opus 4.6	anthropic/claude/opus-4.6	ar, bn, de, en, es, fr, hi, id, it, ja, ko, pt, ru, sw, yo, zh
Anthropic Claude Opus 4.7	anthropic/claude/opus-4.7	ar, bn, de, en, es, fr, hi, id, it, ja, ko, pt, ru, sw, yo, zh
Anthropic Claude Opus 4.8	anthropic/claude/opus-4.8	ar, bn, de, en, es, fr, hi, id, it, ja, ko, pt, ru, sw, yo, zh
OpenAI GPT-5.5	openai/gpt/5.5	am, ar, bg, bn, bs, ca, cs, da, de, el, en, es, et, fa, fi, fr, fr-CA, gu, he, hi, hr, hu, hy, id, is, it, ja, ka, kk, kn, ko, lt, lv, mk, ml, mn, mr, ms, my, nl, no, pa, pl, pt, pt-BR, ro, ru, sk, sl, so, sq, sr, sv, sw, ta, te, tg, th, tl, tr, uk, ur, vi, zh, zh-TW

Reasoning Models #

High-latency but extremely high quality output.

Model	Model ID	Supported Languages

Standard LLM Parameters #

The chat completion endpoint supports standard LLM parameters that modify the behavior of the model:

• frequency_penalty
• presence_penalty
• max_completion_tokens (legacy alias: max_tokens)
• reasoning_effort (for reasoning models only: low, medium, high)
• verbosity (minimal, low, medium, high)
• temperature
• top_p
• logprobs / top_logprobs

See OpenAI’s docs for more information about these parameters.

Tools and LLM Passthrough Parameters #

The following parameters are forwarded to the target LLM and can affect model behavior. Using them bypasses the prose completion cache:

• tools: OpenAI-compatible function tool definitions. Passed through as declaration-only tools — the client executes them and returns role: tool messages. Reserved Agent tool names that cannot be overridden: webSearch, webExtract.
• tool_choice: none, auto, required, or a specific tool object
• logprobs / top_logprobs
• seed, logit_bias, parallel_tool_calls, stop

Compatibility-Only Passthrough Parameters #

These are accepted for OpenAI client compatibility but do not meaningfully change Agent behavior today:

• audio
• modalities
• n (Agent returns a single response)
• prediction
• service_tier
• store
• stream_options

Agent does not currently support multi-modal input or output. Tool definitions are declaration-only — Agent does not execute arbitrary client-supplied tools server-side.

Starter Chatbot App in NextJS #

Check out our bare bones chatbot starter kit which demonstrates how to quickly get up and running with this endpoint using Vercel’s AI SDK. Here is where the magic happens in the endpoint that streams responses from our Agent (chat completions) API:

import { createOpenAICompatible } from "@ai-sdk/openai-compatible";
import { streamText } from "ai";

export async function POST(req: Request) {

  const { messages } = await req.json();

  const apologist = createOpenAICompatible({
    name: 'apologist',
    apiKey: process.env.APOLOGIST_API_KEY,
    baseURL: `${process.env.APOLOGIST_API_URL}`,
  });

  const result = streamText({
    model: apologist('openai/gpt/4o'),
    messages: messages,
  });

  return result.toDataStreamResponse();

}

Yes, it’s really that simple.

Caching #

The completion endpoint supports caching. Under the hood, each version of an Agent’s configuration is stored. In order to have a cached value match, the Agent configuration version, messages (including the current prompt), language, Bible translation, and related request context must all match. The cache is bypassed for json_schema responses and when LLM passthrough parameters such as tools, tool_choice, logprobs, or seed are present. Optional cache TTL may be set via Agent configuration, the x-cache-ttl header, or the cache_ttl query parameter. The cache is very unlikely to be utilized in a chat context, except for pre-defined preliminary prompts. A more common use case would be caching responses for pre-defined text (e.g., a Bible verse).

Learn more about how caching works in our system.

Feedback #

There are 3 additional feedback fields that can set on a given chat completion once it’s been created. These can be used to identify positive, negative, and qualitative feedback on the prompt itself for later viewing / retrieval.

Likes #

Use the POST /api/v1/chat/completions/[prompt_id]/like with a payload as follows to set the liked field:

{
    "liked": [true|false]
}

Flags #

Use the POST /api/v1/chat/completions/[prompt_id]/flag with a payload as follows to set the flagged field:

{
    "flagged": [true|false]
}

Feedback #

Use the POST /api/v1/chat/completions/[prompt_id]/feedback with a payload as follows to set the feedback field:

{
    "feedback": [string|null]
}

Share #

Use POST /api/v1/chat/completions/{prompt_id}/share to create a share record for a completion. All fields are optional:

{
    "conversation_id": "conv-123",
    "session_id": "sess-456",
    "user_id": "user-789"
}

Public shared conversation messages can be retrieved with GET /api/v1/shares/{token} (unauthenticated). List and get completion responses may also include metadata.share_token and metadata.share_url when a conversation is present.

List Chat Completions #

Use the GET /api/v1/chat/completions endpoint to get a paginated list of chat completion records belonging to the authenticated Agent, ordered by prompted_at descending.

Pagination #

Parameter	Type	Default	Max	Description
page	integer	1	—	Page number
per_page	integer	50	100	Records per page

Filter Parameters #

All filters are optional and combinable. Omitting a filter returns all records.

Parameter	Type	Description
language	string	Language code (e.g. en)
min_timestamp	string	Minimum prompted_at (ISO 8601 timestamp)
max_timestamp	string	Maximum prompted_at (ISO 8601 timestamp)
config_id	integer	Agent configuration ID
flagged	boolean	true or false
liked	boolean	true or false
favorited	boolean	true or false
cached	boolean	true or false
conversation_id	string	Conversation identifier
session_id	string	Session identifier
device_id	string	Device identifier
user_id	string	User identifier
bible_id	integer	Bible ID
client	string	Client type (e.g. standalone, embedded, api, channel, benchmark)
channel_id	integer	Channel that submitted the request

Boolean filter parameters accept true/false as strings or native booleans.

Response #

{
    "data": [ /* array of prompt objects */ ],
    "total": 142,
    "page": 1,
    "per_page": 50
}

total reflects the full count matching the applied filters (not just the current page), and can be used to calculate page counts.

---

Get Chat Completion #

Use the GET /api/v1/chat/completions/{prompt_id} endpoint to get a single chat completion record by ID. The endpoint accepts both UUID and legacy numeric integer IDs:

• UUID (current format)
• Numeric integer (legacy)

If the record does not exist or does not belong to the authenticated Agent, a 404 is returned.

Response #

{
    "data": { /* prompt object */ }
}

Prompt Object #

Completion records are returned with the public fields below. The id field is the completion UUID. List/get responses are also enriched with applied tags, top-level guardrail and cta (from response metadata), and share metadata. Get-by-id additionally includes automations.

Field	Type	Description
id	string (UUID)	Unique identifier
language	string	Language code
prompt	string	The user’s prompt text
response	string | null	The generated response text
prompted_at	string	UTC timestamp when the prompt was submitted
response_started_at	string | null	UTC timestamp when streaming began
response_completed_at	string | null	UTC timestamp when the response finished
flagged	boolean	Whether the completion has been flagged
liked	boolean	Whether the completion has been liked
favorited	boolean	Whether the completion has been favorited
cached	boolean	Whether the response was served from cache
feedback	string | null	Free-text feedback submitted by the user
score	integer | null	Numeric score
config_id	integer | null	Agent configuration used
agent_id	integer | null	Agent that handled the completion
agent_token_id	integer | null	API token used to submit the request
channel_id	integer | null	Channel that submitted the request
responder_id	integer | null	Responder used for the completion
performed_web_search	boolean	Whether a web search was performed
tags	array	Applied tags as { id, name } objects
guardrail	object | null	Guardrail metadata when a guardrail fired
cta	object | null	CTA metadata when a CTA was triggered
metadata	object	Share fields such as share_token and share_url
automations	array | null	Automation evaluation results (get-by-id only)
bible_id	integer | null	Bible context used
client	string | null	Client type (standalone, embedded, api, integration)
conversation_id	string | null	Conversation grouping identifier
session_id	string | null	Session identifier
device_id	string | null	Device identifier
user_id	string | null	User identifier
prompt_tokens	integer | null	Token count for the prompt
response_tokens	integer | null	Token count for the response
chat_tokens	integer | null	Token count for the conversation history
reasoning_tokens	integer | null	Token count for reasoning (where applicable)
translated_prompt	string | null	Real-time translated prompt text
translated_response	string | null	Real-time translated response text
reasoning	string | null	Model reasoning output
has_persisted_sources	boolean	Whether source documents were stored
response_metadata	object | null	Additional metadata from the model response
notes	string | null	Internal notes
archived_at	string | null	UTC timestamp when the record was archived

---

Error Responses #

Status	Meaning
400	Invalid JSON in request body
403	Missing or invalid API key
404	Record not found or not accessible by this Agent
422	Validation error (e.g. missing json_schema.schema)
500	Internal server error
503	Agent not found or inactive for the request’s host