24.03.10

Features

• Add support for setting a timeout when pulling Docker images and upgrade aiodocker to version 0.23.0. (#2852)
• Allow DataLoaderManager to get a loader function by function itself rather than function name. (#2717)
• Add an explicit configuration scaling-group-type to agent.toml so that the agent could distinguish whether itself belongs to an SFTP resource group or not (#2796)

Improvements

• Avoid using collections.OrderedDict when not necessary in the manager API and client SDK (#2842)

Fixes

• Merge kernels.role into sessions.session_type and check the image compatibility based on comparison with the ai.backend.role label (#1587)
• Delete vfolder invitation and permission rows when deleting vfolders. (#2780)
• Fix kernel_id assignment for main kernel log retrieval (#2820)
• Wrong count of concurrent compute sessions. (#2829)
• Create kernels with correct scaling_group value. (#2837)
• Fix a regression in progress bar rendering of the TUI installer after upgrading the Textual library (#2867)
• Add scaling_group.agent_count_by_status and scaling_group.agent_total_resource_slots_by_status GQL fields to query the count and the resource allocation of agents that belong to a scaling group. (#2254)
• Fix handling of undefined values in the ModifyImage GraphQL mutation. (#2028)
• Silence model_ namespace warnings with pydantic-based model classes (#2765)
• Change the initialization order of PackageContext to apply target_path correctly in the TUI installer (#2768)
• Make the regex patterns to update configuration files working with multiline texts correctly in the TUI installer (#2771)
• Omit null parameter when call usage-per-period API. (#2777)
• Handle container port mismatch when creating kernel. (#2786)
• Explicitly set the protected service ports depending on the resource group type and the service types (#2797)
• Correct session status determiner function. (#2803)
• Fix endpoint_list.total_count GQL field returning incorrect value (#2805)

External Dependency Updates

• Upgrade Python (3.12.4 -> 3.12.6) and common/tool dependencies to prepare for Python 3.13 and apply latest fixes (#2851)

Miscellaneous

• Enhacne type hints for potential None arguments (#2580)
• Upgrade readthedocs build environment to Python 3.12 (#2814)

Full Changelog

Check out the full changelog until this release (24.03.10).

Full Commit Logs

Check out the full commit logs between release (24.03.10rc1) and (24.03.10).

24.03.10rc1

Features

• Add support for setting a timeout when pulling Docker images and upgrade aiodocker to version 0.23.0. (#2852)

Improvements

• Avoid using collections.OrderedDict when not necessary in the manager API and client SDK (#2842)

Fixes

• Merge kernels.role into sessions.session_type and check the image compatibility based on comparison with the ai.backend.role label (#1587)
• Delete vfolder invitation and permission rows when deleting vfolders. (#2780)
• Fix kernel_id assignment for main kernel log retrieval (#2820)
• Wrong count of concurrent compute sessions. (#2829)
• Create kernels with correct scaling_group value. (#2837)
• Fix a regression in progress bar rendering of the TUI installer after upgrading the Textual library (#2867)

External Dependency Updates

• Upgrade Python (3.12.4 -> 3.12.6) and common/tool dependencies to prepare for Python 3.13 and apply latest fixes (#2851)

Miscellaneous

• Enhacne type hints for potential None arguments (#2580)
• Upgrade readthedocs build environment to Python 3.12 (#2814)

Full Changelog

Check out the full changelog until this release (24.03.10rc1).

Full Commit Logs

Check out the full commit logs between release (24.03.10b3) and (24.03.10rc1).

24.03.10b3

No significant changes.

Full Changelog

Check out the full changelog until this release (24.03.10b3).

Full Commit Logs

Check out the full commit logs between release (24.03.10b2) and (24.03.10b3).

24.03.10b2

Fixes

• Fix Service.create() SDK method and service create CLI command not working with UnboundLocalError exception (#2806)

Full Changelog

Check out the full changelog until this release (24.03.10b2).

Full Commit Logs

Check out the full commit logs between release (24.03.10b1) and (24.03.10b2).

24.03.10b1

Features

• Allow DataLoaderManager to get a loader function by function itself rather than function name. (#2717)
• Add an explicit configuration scaling-group-type to agent.toml so that the agent could distinguish whether itself belongs to an SFTP resource group or not (#2796)

Fixes

• Fix handling of undefined values in the ModifyImage GraphQL mutation. (#2028)
• Silence model_ namespace warnings with pydantic-based model classes (#2765)
• Change the initialization order of PackageContext to apply target_path correctly in the TUI installer (#2768)
• Make the regex patterns to update configuration files working with multiline texts correctly in the TUI installer (#2771)
• Omit null parameter when call usage-per-period API. (#2777)
• Handle container port mismatch when creating kernel. (#2786)
• Explicitly set the protected service ports depending on the resource group type and the service types (#2797)
• Correct session status determiner function. (#2803)
• Fix endpoint_list.total_count GQL field returning incorrect value (#2805)

Full Changelog

Check out the full changelog until this release (24.03.10b1).

Full Commit Logs

Check out the full commit logs between release (24.03.9) and (24.03.10b1).

24.03.9

Features

• Allow filter and order in endpointlist gql request. (#2723)
• Add scaling_group.agent_count_by_status and scaling_group.agent_total_resource_slots_by_status GQL fields to query the count and the resource allocation of agents that belong to a scaling group. (#2254)
• Allow bulk association and disassociation of scaling groups with domains, user groups, and key pairs. (#2473)
• Add new vfolder API to update sharing status. (#2740)

Improvements

• Enable robust DB connection handling by allowing pool-pre-ping setting. (#1991)

Fixes

• Correct msgpack deserialization of ResourceSlot. (#2754)
• Fix regression error of session create_from_template command. (#2761)
• Fix GET /func/folders/{folderName} API returning string literal "null" instead of null value on user and group fields (#2584)
• Fix list_files, get_fstab_contents, get_performance_metric and shared_vfolder_info Python SDK function not working with ValidationError exception printed (#2706)
• Handle OS Error when deleting vfolders. (#2741)
• Fix typo in Virtual-folder status update code. (#2742)

Full Changelog

Check out the full changelog until this release (24.03.9).

Full Commit Logs

Check out the full commit logs between release (24.03.9rc1) and (24.03.9).

24.03.9rc1

Features

• Allow filter and order in endpointlist gql request. (#2723)

Fixes

• Correct msgpack deserialization of ResourceSlot. (#2754)
• Fix regression error of session create_from_template command. (#2761)

Full Changelog

Check out the full changelog until this release (24.03.9rc1).

Full Commit Logs

Check out the full commit logs between release (24.03.9b1) and (24.03.9rc1).

24.03.9b1

Features

• Add scaling_group.agent_count_by_status and scaling_group.agent_total_resource_slots_by_status GQL fields to query the count and the resource allocation of agents that belong to a scaling group. (#2254)
• Allow bulk association and disassociation of scaling groups with domains, user groups, and key pairs. (#2473)
• Add new vfolder API to update sharing status. (#2740)

Improvements

• Enable robust DB connection handling by allowing pool-pre-ping setting. (#1991)

Fixes

• Fix GET /func/folders/{folderName} API returning string literal "null" instead of null value on user and group fields (#2584)
• Fix list_files, get_fstab_contents, get_performance_metric and shared_vfolder_info Python SDK function not working with ValidationError exception printed (#2706)
• Handle OS Error when deleting vfolders. (#2741)
• Fix typo in Virtual-folder status update code. (#2742)

Full Changelog

Check out the full changelog until this release (24.03.9b1).

Full Commit Logs

Check out the full commit logs between release (24.03.8) and (24.03.9b1).

24.03.8

Features

• Add an enable_LLM_playground option to show/hide the LLM playground tab on the serving page. (#2677)
• Add max_atom_plus_device_per_container config on webserver (#2686)

Fixes

• Remove duplicate CPU quota arguments when creating containers (#2608)
• Allow sudo-enabled container users to ovewrite /usr/bin/scp and /usr/libexec/sftp-server by unifying the intrinsic ssh binaries to use the merged dropbearmulti executable. (#2671)
• Update webserver logout API to respond with HTTP 200 OK (#2681)
• Scan parent directory of created qtree to avoid creating quota on non-existing directory. (#2696)

Full Changelog

Check out the full changelog until this release (24.03.8).

Full Commit Logs

Check out the full commit logs between release (24.03.8rc2) and (24.03.8).

24.03.8rc2

Fixes

• skip resolving malformed ModelCard GQL item (#2570)
• Robust type check when idle checker fetches utilization data. (#2601)
• Let GET /resource/usage/period request contain data in query parameter rather than JSON body. (#2661)

Full Changelog

Check out the full changelog until this release (24.03.8rc2).

Full Commit Logs

Check out the full commit logs between release (24.03.8rc1) and (24.03.8rc2).