Commit Graph

63 Commits

Author SHA1 Message Date
LucasArmandVast 4380d98c01 Use PyWorker SDK (#67)
* Change PyWorker to Worker SDK
* Moved /lib to vast-sdk (https://github.com/vast-ai/vast-sdk)
2025-12-15 19:33:03 -08:00
Colter Downing 222ac2a0dd default endpoint name 2025-12-04 10:54:55 -08:00
Colter Downing 40aed9b5f8 adding s3 as an option 2025-12-04 10:52:57 -08:00
Colter Downing d4d36bf86e done with comfy updates 2025-12-03 20:45:55 -08:00
Colter Downing e839cfc6e8 include view in API wrapper 2025-12-03 20:22:45 -08:00
Colter Downing f04138e13b update to be able to get images 2025-12-03 20:16:25 -08:00
Colter Downing 6b5b1341a7 update tgi client 2025-12-03 18:38:42 -08:00
Colter-Downing 8be92c03de Merge pull request #69 from vast-ai/AUTO-874--fix-openai-worker-client
defaults to ENDPOINT_NAME and DEFAULT_MODEL but uses the flag first
2025-12-03 16:59:56 -08:00
Colter Downing adedb8ba90 defaults to ENDPOINT_NAME and DEFAULT_MODEL but uses the flag first if present 2025-12-03 16:57:28 -08:00
Lucas Armand 0bcd2219ea Increase model wait time for vLLM 2025-12-03 12:38:52 -08:00
Lucas Armand e0449cb3c7 add llama log 2025-11-21 10:22:16 -08:00
Lucas Armand 3adec1826d minor changes 2025-11-11 17:11:38 -08:00
Lucas Armand b55bfa9611 Updated clients, include vastai-sdk, handle non-UTF-8 2025-11-11 17:09:28 -08:00
Abiola Akinnubi 2cde573c56 Merge pull request #48 from vast-ai/comfy-request-idx
Added request_idx to comfy auth_data
2025-10-30 11:27:35 -07:00
Abiola Akinnubi 944f83fc03 Removed extra spaces from operator assignment 2025-10-28 21:03:52 +00:00
LucasArmandVast d6a6e34c6b Merge branch 'main' into new-pyworker 2025-10-27 12:43:49 -07:00
Abiola Akinnubi f56bbc0ebe Added request_idx to comfy auth_data 2025-10-27 03:17:06 +00:00
Colter Downing bcecd6df40 Suppress matplot debug logs 2025-10-25 16:18:02 -07:00
Rob Ballantyne f4f7080df1 Re-add comment 2025-10-23 17:00:28 +01:00
Rob Ballantyne d51a338e8f log when benchmark file not used 2025-10-23 16:41:02 +01:00
Rob Ballantyne 92a04bd7af No silent fail if benchmark file is missing 2025-10-23 13:41:03 +01:00
Rob Ballantyne ec25dda3ad Merge branch 'vast-ai:main' into feat/comfyui-json-benchmark-workflow-from-file 2025-10-08 14:49:32 +01:00
Rob Ballantyne 4fdc314fd9 Fix healthcheck endpoint URL 2025-10-06 22:16:09 +01:00
Rob Ballantyne 3786cf978d Add awareness of errors thrown by the provisioning script 2025-10-05 23:14:59 +01:00
Rob Ballantyne a86d4bcf9c Import json 2025-10-05 23:05:33 +01:00
Rob Ballantyne e9b6a14a5e Import Path 2025-10-05 22:59:19 +01:00
Rob Ballantyne cadac033e1 Enables use of custom workflow for benchmarking
Retains existing method is misc/benchmark.json is nopt present
2025-10-05 22:53:22 +01:00
abiola-vastai 38782d89bc undo the fix for comfy yesterday. 2025-09-03 17:12:35 +00:00
abiola-vastai b20d9e714c Blind hotfix to see if comfy UI default is needed. if it does work we would revert back. 2025-09-03 01:20:09 +00:00
Rob Ballantyne b8377c4081 Set cost to 100 2025-08-28 16:13:17 +01:00
Rob Ballantyne 703435d10e Improve MODEL_SERVER_START_* messages 2025-08-26 12:42:04 +01:00
Rob Ballantyne 947fc5eea4 Improve benchmarking explanation 2025-08-26 12:41:30 +01:00
Rob Ballantyne 7c1a544b19 Improve error reporting when no ready workers 2025-08-26 12:41:05 +01:00
Rob Ballantyne 16b414676e Use count_workload() function for cost 2025-08-25 18:31:10 +01:00
Rob Ballantyne ba74ac8136 Use cost value 1 for all jobs 2025-08-25 17:58:22 +01:00
Rob Ballantyne 92ff412679 Use MODEL_SERVER_URL environment variable 2025-08-25 17:57:32 +01:00
Rob Ballantyne fc75a64684 Use MODEL_SERVER_URL environment variable 2025-08-25 17:56:27 +01:00
Rob Ballantyne 3f4acb29fa Improved client exception handling 2025-08-22 15:20:15 +01:00
Rob Ballantyne 58b078f908 Fix modifier class 2025-08-20 18:06:02 +01:00
Rob Ballantyne f9fdf04884 Fix signature 2025-08-20 13:27:29 +01:00
Rob Ballantyne 636f17d27f Fix workflow modifier class 2025-08-20 09:57:07 +01:00
Rob Ballantyne 8797b504af Initial ComfyUI implementation with updated wrapper 2025-08-19 17:59:20 +01:00
Nader Arbabian 0bf3247a34 fix completions and interactive client 2025-08-11 12:37:53 -07:00
Nader Arbabian 8804e17201 download vast.ai's root certificate in order to make pyworker requests (#25) 2025-08-08 17:04:16 -07:00
Rob Ballantyne e0be45f39a Addresses breaking change in core pyworker (#22)
* Addresses breaking change in test_utils.py

Endpoint.get_endpoint_api_key() now requires instance

Moves the call to this function out of the APIClient and into main

* Ensure make_benchmark_payload has a value to calculate the workload

---------

Co-authored-by: Nader Arbabian <nader@vast.ai>
2025-07-18 16:11:10 -07:00
Nader Arbabian be2aafdb1f fix pyright errors + revert to old way of handling cancelled api requests (#23) 2025-07-17 16:59:06 -07:00
Rob Ballantyne 69d9b7455f OpenAI compatible worker (#19)
Adds initial support for OpenAI compatible inference servers

Available endpoints:

- `/v1/completions`
- `/v1/chat/completions`
2025-07-16 09:46:26 +01:00
Nader Arbabian 6fb610cb5b fix pyworker miscounting active connections (#20)
* fix pyworker miscounting active connections

* clean up some issues

* add option to skip auth
2025-07-15 15:33:27 -07:00
Nader Arbabian 4bac805093 update tokenizers 2025-06-10 17:07:38 -07:00
Abiola Akinnubi 71ed54ebe4 Endpoint update pr one (#1)
* Added endpoint flexibility along with existing log. extended the log support

* Switched Endpoint back to vast-ai, Added endpoint flexibility along with existing log. extended the log support

* Modify the endpoint return type as optional and check via pyright to ensure there are not compilation/type errors

* Added endpoint flexibility along with existing log. extended the log support

Switched Endpoint back to vast-ai, Added endpoint flexibility along with existing log. extended the log support

Modify the endpoint return type as optional and check via pyright to ensure there are not compilation/type errors

* Endpoint Utils and API changes
2025-06-02 18:43:27 -07:00