examples/ollama-local: integrate Pydantic AI with a local Ollama backend for on-prem semantic layering
Python · active 2025-03-29 → 2026-08-10 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 500 days from 2025-03-29 to 2026-08-10. Pushes: 3,611 total, peak 54 in a day. Pull requests: 2,331 total, peak 30 in a day. Issues: 1,725 total, peak 21 in a day. Comments: 10,293 total, peak 134 in a day. Stars: 3,869 total, peak 74 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| DouweM | 7.1K | 1.8K | 648 | 3.3K |
| github-actions[bot] | 2.6K | 0 | 561 | 1.3K |
| devin-ai-integration[bot] | 2K | 0 | 0 | 944 |
| dsfaccini | 1.6K | 534 | 172 | 455 |
| Kludex | 1.2K | 357 | 197 | 374 |
| dmontagu | 805 | 323 | 71 | 202 |
| adtyavrdhn | 679 | 312 | 27 | 180 |
| alexmojaki | 248 | 100 | 30 | 72 |
| samuelcolvin | 186 | 59 | 32 | 71 |
| Copilot | 135 | 0 | 2 | 73 |
| mattbrandman | 119 | 0 | 4 | 78 |
| brightsparc | 88 | 0 | 0 | 45 |
| hyperlint-ai[bot] | 86 | 0 | 0 | 86 |
| Danipulok | 81 | 0 | 7 | 45 |
| Wh1isper | 81 | 0 | 8 | 52 |
| strawgate | 79 | 16 | 7 | 42 |
| DougTrajano | 78 | 0 | 2 | 47 |
| pydanty[bot] | 78 | 5 | 4 | 12 |
| fswair | 73 | 0 | 8 | 56 |
| Viicos | 67 | 19 | 12 | 25 |
Recent activity
Latest issues, pull requests and releases
- Issue#4262github-actions[bot]2026-08-09 00:36[RFC] Strict Runtime Validation for Untrusted Tool Outputs
- Issue comment#7249dgvj-work2026-08-08 03:51Export `HistoryProcessor` from the package root
- Issue comment#7159dsfaccini2026-08-08 03:08gzip response bodies are silently truncated: multi-member streams lose members, cut-off streams are accepted
- Issue#7159dsfaccini2026-08-08 03:08gzip response bodies are silently truncated: multi-member streams lose members, cut-off streams are accepted
- Issue#7159dsfaccini2026-08-08 03:08gzip response bodies are silently truncated: multi-member streams lose members, cut-off streams are accepted
- Issue#6792dsfaccini2026-08-08 02:54`GoogleModel` provider detection is name-based, so `GoogleProvider(client=genai.Client(vertexai=True))` is misrouted as the Gemini API
- Issue comment#6067leadingproblemsolver2026-08-07 19:43Expose the live `AgentRun` handle from `to_web`/`dispatch_request` for cross-request steering (interrupt / enqueue)
- Pull request#7287DouweM2026-08-07 19:43
- Issue comment#6974leadingproblemsolver2026-08-07 19:36Temporal replay re-emits every Pydantic AI span, inflating telemetry 6-13x
- Pull request#7285DouweM2026-08-07 19:31
- Pull request#7271github-actions[bot]2026-08-07 19:28
- Pull request#7286DouweM2026-08-07 19:24
- Issue#7128DouweM2026-08-07 19:20Add openai_cache_instructions: place prompt_cache_breakpoint on the developer/system message for OpenAI GPT-5.6
- Pull request#7285github-actions[bot]2026-08-07 19:17
- Pull request#7266DouweM2026-08-07 17:56
- Issue#7284pydanty[bot]2026-08-07 17:23`OpenAIEmbeddingModel.embed()` doesn't preserve batch response order (drops per-item index)
- Issue comment#7284pydanty[bot]2026-08-07 17:11`OpenAIEmbeddingModel.embed()` doesn't preserve batch response order (drops per-item index)
- Issue comment#6571DouweM2026-08-07 16:51feat: distinguish no-output retries from validation feedback in `RetryPromptPart`
- Issue comment#7277dsfaccini2026-08-07 16:37Document `run_sync()` as top-level-only when delegating to another agent
- Pull request#7248pydanty[bot]2026-08-07 02:49
- Issue#7171pydanty[bot]2026-08-05 10:02ToolOutput validation retries duplicate the full input once per error, causing multiplicative context growth
- Pull request#7138dsfaccini2026-08-04 18:26
- Pull request#2684github-actions[bot]2026-07-31 18:41
- Issue#7004pydanty[bot]2026-07-31 18:22`UserError`: `CachePoint cannot be the first content in a user message` when using `Planning` capability via AWS Bedrock
- Issue comment#6989macroscopeapp[bot]2026-07-31 01:14Fix Temporal `ActivityConfig` validation on Python 3.10 and 3.11
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 3,869 stars here means stars gained during the window, not the repo's star count.