Generate API (Ollama Compatible)¶

Generate text for a single prompt with Amazon Bedrock models through the Ollama /api/generate interface. Served under /ollama by default; the examples below use $BASE, which includes that prefix.

At a glance¶

• Drop-in Ollama compatibility — Follows the Ollama /api/generate request and response shape, including its newline-delimited JSON streaming transport, so an existing Ollama client works by changing the base URL.
• Single-prompt simplicity — A minimal shape for clients that send one prompt at a time rather than a full conversation.
• Structured output — format accepts "json" or a full JSON Schema, and the answer is constrained to it.
• Private AWS backend — Served entirely by Amazon Bedrock models in your own AWS account — no traffic to third-party endpoints.
• Image input beyond base64 — images also takes a URL, a data URI or an s3:// URI on models that read images.
• Differs from the Ollama API: raw, suffix, template and context are refused with 400 and context is never returned; models are never resident, so keep_alive holds nothing loaded — see Limits and behaviour to know.

export BASE="https://your-host/ollama"  # <scheme>://<host> + OLLAMA_ROUTES_PREFIX

curl -X POST "$BASE/api/generate" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "amazon.nova-micro-v1:0",
    "prompt": "Why is the sky blue?",
    "stream": false
  }'

Endpoints¶

/api/generate is served in full; new integrations usually pick /api/chat, which carries message history, tool calls and multi-turn thinking.

Model Names¶

Send the model names GET /api/tags publishes — those are the canonical identifiers this server resolves directly. A trailing :latest is accepted and stripped as a fallback when the exact name is not found. Short aliases accepted on this server's other APIs also work here even though /api/tags does not list them. A name learned from ollama.com (for example llama3.2:3b) is not available through this server and answers 404. Every response echoes the model name exactly as the request spelled it, in model.

Find compatible models: Call /search_models with route=ollama_generate to discover model IDs that support this route, or call GET /api/tags for the Ollama-shaped listing.

Feature compatibility¶

Streaming¶

By default (stream unset, or true), the response body is newline-delimited JSON (application/x-ndjson): one bare JSON object per line, with no data: prefix and no [DONE] sentinel. The stream is terminated by an object carrying "done": true and the response metrics. See Streaming on the Chat API for the transport details and how a mid-stream failure is reported.

curl -N -X POST "$BASE/api/generate" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "amazon.nova-micro-v1:0",
    "prompt": "Write a haiku about the sea."
  }'

{"model":"amazon.nova-micro-v1:0","created_at":"2026-08-27T12:00:00.000000+00:00","response":"Waves","done":false}
{"model":"amazon.nova-micro-v1:0","created_at":"2026-08-27T12:00:00.100000+00:00","response":" crash on silent shores","done":false}
{"model":"amazon.nova-micro-v1:0","created_at":"2026-08-27T12:00:00.200000+00:00","response":"","done":true,"done_reason":"stop","total_duration":812345678,"prompt_eval_count":9,"prompt_eval_duration":123456789,"eval_count":11,"eval_duration":688888889}

Fields Not Available¶

raw, suffix, template and context are each rejected with 400 when set. Every one of them needs the model's own prompt template or tokenizer to honor — the exact text the model was trained to expect around a prompt, or the token IDs of a prior turn — and neither is exposed here. The error names the offending field and points at /api/chat as the alternative. Because there is no token context to hand back, response.context is never returned by this endpoint.

Send prompt together with system for a system-prompted single turn, or use /api/chat for a multi-turn conversation.

Images¶

images accepts base64-encoded image data, as Ollama does. This server additionally accepts a URL, a data URI or an s3:// URI in the same field, on models that support image input.

Loading and Unloading¶

A request with no prompt is upstream's way of making a model resident — it is what ollama run <model> sends before it opens its REPL — and the same request with keep_alive set to 0 is what ollama stop <model> sends to evict it. A hosted model is always resident, so both are answered without invoking anything:

{
  "model": "amazon.nova-micro-v1:0",
  "created_at": "2026-01-01T00:00:00+00:00",
  "response": "",
  "done": true,
  "done_reason": "load"
}

done_reason is unload when keep_alive is 0, load otherwise. The answer is a single JSON object whatever stream says, as upstream's is.

Limits and behaviour to know¶

• raw, suffix, template and context are rejected with 400 — see Fields Not Available.
• logprobs and top_logprobs are rejected with 400 — log probabilities are not available.
• keep_alive does not keep anything loaded: models are never resident. It is read only to tell a prompt-less request's done_reason apart, load from unload.
• Runner options inside options (num_ctx, num_gpu, num_thread, num_batch, main_gpu, use_mmap, min_p, and any other key a local runner would use) are accepted and ignored.
• load_duration is never reported: there is no model-loading phase to measure, and a number there would be invented.
• prompt_eval_duration and eval_duration are reported only when streaming. All duration and count fields are optional in the Ollama API, so a client computing tokens-per-second from a non-streamed response has no duration to divide by.
• A model name learned from ollama.com — llama3.2:3b, for one — names nothing this server serves and answers 404; send a name GET /api/tags publishes.

Request headers¶

A local Ollama server needs no key; this one does, on every route.

Try it¶

curl -X POST "$BASE/api/generate" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "amazon.nova-micro-v1:0",
    "prompt": "Why is the sky blue?",
    "stream": false
  }'

Example response:

{
  "model": "amazon.nova-micro-v1:0",
  "created_at": "2026-08-27T12:00:00.000000+00:00",
  "response": "The sky is blue because of Rayleigh scattering...",
  "done": true,
  "done_reason": "stop",
  "total_duration": 734567890,
  "prompt_eval_count": 7,
  "eval_count": 42
}

Next steps¶

Next: Chat API · Embed API · Models API · Search Models API