)

* ci: regenerated with OpenAPI Doc , Speakeasy CLI 1.685.0

* Fixing the version in pyproject.toml and removing opentelemetry-semantic-conventions

---------

Co-authored-by: speakeasybot <bot@speakeasyapi.dev>
Co-authored-by: Guillaume Dumont <guillaume.dumont@mistral.ai>

)

* ci: regenerated with OpenAPI Doc , Speakeasy CLI 1.685.0

* Removing opentelemetry-semantic-conventions

---------

Co-authored-by: speakeasybot <bot@speakeasyapi.dev>
Co-authored-by: Guillaume Dumont <guillaume.dumont@mistral.ai>

* Bump version from 1.12.1 to 1.12.2

This change was omitted during the last version bump since it is not taken care of my the Speakeasy CI (file is in .genignore).

* Update locks

* Readding code samples

)

* ## Python SDK Changes:
* `mistral.fine-tuning.jobs.create()`: 
  *  `request` **Changed** **Breaking** ⚠️
  *  `response` **Changed** **Breaking** ⚠️
* `mistral.models.update()`:  `response` **Changed** **Breaking** ⚠️
* `mistral.models.archive()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.models.unarchive()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.batch.jobs.cancel()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.batch.jobs.get()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.batch.jobs.create()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.batch.jobs.list()`: 
  *  `request.order_by` **Added**
  *  `response` **Changed** **Breaking** ⚠️
* `mistral.fine-tuning.jobs.start()`:  `response` **Changed** **Breaking** ⚠️
* `mistral.fine-tuning.jobs.cancel()`:  `response` **Changed** **Breaking** ⚠️
* `mistral.fine-tuning.jobs.get()`:  `response` **Changed** **Breaking** ⚠️
* `mistral.fine-tuning.jobs.list()`:  `response` **Changed** **Breaking** ⚠️
* `mistral.beta.agents.list()`: 
  *  `request.search` **Added**
  *  `response.[].version_message` **Added**
* `mistral.beta.agents.get_version()`:  `response.version_message` **Added**
* `mistral.beta.agents.list_versions()`:  `response.[].version_message` **Added**
* `mistral.beta.agents.update_version()`:  `response.version_message` **Added**
* `mistral.beta.agents.update()`: 
  *  `request.version_message` **Added**
  *  `response.version_message` **Added**
* `mistral.beta.agents.get()`:  `response.version_message` **Added**
* `mistral.beta.agents.delete_version_alias()`: **Added**
* `mistral.beta.agents.create()`: 
  *  `request.version_message` **Added**
  *  `response.version_message` **Added**

* ci

* fix: run on v1

* feat(ci): add manual confirmation for v2 SDK publishing

This change introduces a workflow modification that requires manual confirmation before publishing the v2 SDK (mistralai.client namespace), which is still in WIP/alpha status. The workflow now also auto-publishes from the v1 branch without requiring manual confirmation. This ensures that only intentional publishing of the v2 SDK occurs while maintaining automated publishing for the stable v1 SDK.

* feat(realtime): add support for target streaming delay in realtime transcription

This commit introduces the ability to specify a target streaming delay in milliseconds for realtime transcription sessions. The feature is implemented across the library and examples, allowing users to control the latency between audio input and transcription output.

The changes include:
- Adding a new parameter to the `transcribe_stream` method and related functions
- Updating the RealtimeConnection class to handle streaming delay updates
- Modifying the example scripts to demonstrate dual-delay transcription
- Adding proper type hints and documentation for the new functionality

The feature is backward compatible, maintaining existing behavior when no delay is specified.

---------

Co-authored-by: speakeasybot <bot@speakeasyapi.dev>
Co-authored-by: jean-malo <jm@mistral.ai>

* fix: 1.12.3

* lock

This commit introduces proper initialization checks for PyAudio in both realtime transcription examples. It adds:

1. A new utility function `load_pyaudio()` to handle PyAudio initialization
2. Error handling for cases where PyAudio cannot be initialized
3. Early termination with error message when PyAudio is not available
4. Consistent error handling across both microphone examples

The changes ensure the applications fail gracefully when PyAudio is not available or cannot be initialized, providing better user feedback.

* fix: change constFieldCasing from upper to normal

Keep const fields lowercase (e.g., `role` instead of `ROLE`) to match
the actual API values and align with the private SDK.

* fix: rm outdated examples and add flush

* chore: update excluded files list in run_examples.sh

Add async_realtime_transcription_dual_delay_microphone.py to the excluded files list in the run_examples.sh script to prevent it from being executed during example runs.

* fix(examples): update excluded job example file name

The script was excluding an incorrect file name ("job.py") in the exclude_files array, which has been updated to the correct name ("jobs.py") to ensure the proper example is skipped during execution.

* fix: typing

* chore(version): bump version to 1.12.4

Update the package version to 1.12.4 in preparation for a new release.

* ## Python SDK Changes: (#365)

* `mistral.beta.conversations.start()`: 
  *  `request.inputs.[array].[]` **Changed** **Breaking** ⚠️
  *  `response` **Changed** **Breaking** ⚠️
* `mistral.beta.conversations.list()`:  `response.[]` **Changed** **Breaking** ⚠️
* `mistral.beta.conversations.get()`:  `response` **Changed** **Breaking** ⚠️
* `mistral.beta.conversations.append()`: 
  *  `request.inputs.[array].[]` **Changed** **Breaking** ⚠️
  *  `response` **Changed** **Breaking** ⚠️
* `mistral.beta.conversations.get_history()`:  `response` **Changed** **Breaking** ⚠️
* `mistral.beta.conversations.get_messages()`:  `response` **Changed** **Breaking** ⚠️
* `mistral.beta.conversations.restart()`: 
  *  `request.inputs.[array].[]` **Changed** **Breaking** ⚠️
  *  `response` **Changed** **Breaking** ⚠️
* `mistral.beta.conversations.start_stream()`: 
  *  `request.inputs.[array].[]` **Changed** **Breaking** ⚠️
  *  `response.[].data` **Changed** **Breaking** ⚠️
* `mistral.beta.conversations.append_stream()`: 
  *  `request.inputs.[array].[]` **Changed** **Breaking** ⚠️
  *  `response.[].data` **Changed** **Breaking** ⚠️
* `mistral.beta.conversations.restart_stream()`: 
  *  `request.inputs.[array].[]` **Changed** **Breaking** ⚠️
  *  `response.[].data` **Changed** **Breaking** ⚠️
* `mistral.beta.agents.create()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.beta.agents.list()`:  `response.[].object` **Changed** **Breaking** ⚠️
* `mistral.beta.agents.get()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.beta.agents.update()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.beta.agents.update_version()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.beta.agents.list_versions()`:  `response.[].object` **Changed** **Breaking** ⚠️
* `mistral.beta.agents.get_version()`:  `response.object` **Changed** **Breaking** ⚠️
* `mistral.chat.complete()`: 
  *  `request.messages.[]` **Changed** **Breaking** ⚠️
  *  `response.choices.[].message` **Changed** **Breaking** ⚠️
* `mistral.chat.stream()`: 
  *  `request.messages.[]` **Changed** **Breaking** ⚠️
  *  `response.[].data.choices.[].delta.content.[array].[]` **Changed** **Breaking** ⚠️
* `mistral.fim.complete()`:  `response.choices.[].message` **Changed** **Breaking** ⚠️
* `mistral.fim.stream()`:  `response.[].data.choices.[].delta.content.[array].[]` **Changed** **Breaking** ⚠️
* `mistral.agents.complete()`: 
  *  `request.messages.[]` **Changed** **Breaking** ⚠️
  *  `response.choices.[].message` **Changed** **Breaking** ⚠️
* `mistral.agents.stream()`: 
  *  `request.messages.[]` **Changed** **Breaking** ⚠️
  *  `response.[].data.choices.[].delta.content.[array].[]` **Changed** **Breaking** ⚠️
* `mistral.classifiers.moderate_chat()`: 
  *  `request.inputs.[array].[]` **Changed** **Breaking** ⚠️
* `mistral.classifiers.classify_chat()`: 
  *  `request.inputs.[inputs].messages.[]` **Changed** **Breaking** ⚠️
* `mistral.ocr.process()`:  `request.document` **Changed** **Breaking** ⚠️
* `mistral.audio.transcriptions.complete()`:  `response.segments.[].type` **Changed** **Breaking** ⚠️
* `mistral.audio.transcriptions.stream()`:  `response.[].data` **Changed** **Breaking** ⚠️

Co-authored-by: speakeasybot <bot@speakeasyapi.dev>

---------

Co-authored-by: jean-malo <jm@mistral.ai>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: speakeasybot <bot@speakeasyapi.dev>