– UNSUPPORTED

pydantic-ai 2.61.0

stream.utf8 · ollama

Requirements

Id	Level	Status	Evidence

Notes

pydantic-ai 2.61.0 implements OllamaModel via OpenAI-compatible Chat Completions only; there is no native /api/chat client. (https://ai.pydantic.dev/models/ollama/)

What the provider sent

Finish: stop
Raw finish: stop
Text: Café Zürich — καλημέρα! 東京で会いましょう 🌸🎉
Tool calls: []
Usage: { "alternatives": {}, "cached_input": 0, "input": 35, "output": 22, "raw": { "eval_count": 22, "prompt_eval_cached_count": 0, "prompt_eval_count": 35 }, "reasoning": null, "total": 57 }
Stream completed: true
Error: null

Wire excerpt

status: 200
content-type: application/x-ndjson
body (ndjson):
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.416188935Z","message":{"role":"assistant","content":"C"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.530078243Z","message":{"role":"assistant","content":"af"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.530111546Z","message":{"role":"assistant","content":"é"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.643562986Z","message":{"role":"assistant","content":" Zürich"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.643838961Z","message":{"role":"assistant","content":" —"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.758817824Z","message":{"role":"assistant","content":" κα"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.758877305Z","message":{"role":"assistant","content":"λη"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.75909952Z","message":{"role":"assistant","content":"μέ"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.87436023Z","message":{"role":"assistant","content":"ρα"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.874416274Z","message":{"role":"assistant","content":"!"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.988950397Z","message":{"role":"assistant","content":" 東京"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.988976405Z","message":{"role":"assistant","content":"で"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:05.989062476Z","message":{"role":"assistant","content":"会い"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:06.10420813Z","message":{"role":"assistant","content":"ましょう"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:06.21906252Z","message":{"role":"assistant","content":" 🌸"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:06.334083943Z","message":{"role":"assistant","content":"🎉"},"done":false}
{"model":"qwen3.5:0.8b","created_at":"2026-10-12T19:01:06.334239784Z","message":{"role":"assistant","content":""},"done":true,"done_reason":"stop","total_duration":1311450038,"load_duration":2172284,"prompt_eval_count":35,"prompt_eval_cached_count":0,"prompt_eval_duration":382649000,"eval_count":22,"eval_duration":917814000}

What the library reported

Outcome

unsupported

Finish

{
  "raw": null,
  "source": null,
  "surface": "absent",
  "value": null
}

Text

Value: null
Surface: absent

Tool calls

[]

Usage

{
  "cached_input": null,
  "input": null,
  "output": null,
  "reasoning": null,
  "surface": {
    "cached_input": "absent",
    "input": "absent",
    "output": "absent",
    "reasoning": "absent",
    "total": "absent"
  },
  "total": null
}

Stream

{
  "chunks_seen": 0,
  "completed": null,
  "used": true
}

Error

null

Unsupported reason

pydantic-ai 2.61.0 implements OllamaModel via OpenAI-compatible Chat Completions only; there is no native /api/chat client. https://ai.pydantic.dev/models/ollama/

Full observation JSON

{
  "elapsed_ms": 1.02215725928545,
  "error": null,
  "finish": {
    "raw": null,
    "source": null,
    "surface": "absent",
    "value": null
  },
  "format": "wiretruth.observation/v1",
  "library": {
    "api": "Agent.run_stream_events",
    "language": "python",
    "name": "pydantic-ai",
    "version": "2.61.0"
  },
  "notes": [],
  "outcome": "unsupported",
  "stream": {
    "chunks_seen": 0,
    "completed": null,
    "used": true
  },
  "structured": {
    "attempted": false,
    "returned_as_success": false,
    "value": null
  },
  "text": {
    "surface": "absent",
    "value": null
  },
  "tool_calls": [],
  "unsupported_reason": "pydantic-ai 2.61.0 implements OllamaModel via OpenAI-compatible Chat Completions only; there is no native /api/chat client. https://ai.pydantic.dev/models/ollama/",
  "usage": {
    "cached_input": null,
    "input": null,
    "output": null,
    "reasoning": null,
    "surface": {
      "cached_input": "absent",
      "input": "absent",
      "output": "absent",
      "reasoning": "absent",
      "total": "absent"
    },
    "total": null
  }
}

Fixture

recorded ollama/stream.utf8

Provenance	Value
Recorded at	2026-10-12T19:01:05Z
Requested model	qwen3.5:0.8b
Returned model	qwen3.5:0.8b
CI run	https://github.com/kartsan03/wiretruth/actions/runs/38294709109

Reproduce

wiretruth serve ollama/stream.utf8

Plain-library snippet from the shim manifest.

Unsupported: OllamaModel uses only the OpenAI-compatible /v1 API, not native /api/chat.

Upstream

No matching upstream report.

History

Version	Grade
2.61.0	– UNSUPPORTED
2.56.0	– UNSUPPORTED