Commit Graph

45 Commits

Author SHA1 Message Date
nai-degen 0c448cb59d fixes azure dalle using wrong rate limit and out-of-spec Retry-After header 2024-08-27 23:53:28 -05:00
nai-degen d706d4c59d adds USER_CONCURRENCY_LIMIT environment variable 2024-06-14 22:52:16 -05:00
nai-degen 7c64d9209e minor refactoring of response middleware handlers 2024-03-17 22:20:39 -05:00
nai-degen 03c5c473e1 improves error handling for sillytavern 2024-03-04 22:59:32 -06:00
nai-degen e813cd9d22 default claude 2.1 instead of 1.3 in openai compat endpoint since 1.3 is not accessible on all keys 2024-01-18 04:14:15 -06:00
nai-degen 7b0892ddae fixes unawaited call to async enqueue 2024-01-07 16:23:53 -06:00
nai-degen 7f92565739 SSE queueing adjustments, untested 2024-01-07 16:19:22 -06:00
nai-degen 8dc7464381 strips extraneous properties on zod schemas 2024-01-07 13:00:48 -06:00
nai-degen 5599a83ae4 improves streaming error handling 2023-12-14 05:01:10 -06:00
nai-degen 94d4efe9bb properly enforce allowedModelFamilies; refactor HPM proxyReq handlers 2023-12-05 22:07:56 -06:00
nai-degen fdd824f0e4 adds azure rate limit auto-retry 2023-12-04 01:24:33 -06:00
khanon fbdea30264 Azure OpenAI suport (khanon/oai-reverse-proxy!48) 2023-12-04 04:21:18 +00:00
khanon f29049f993 Support for GPT-4-Vision (khanon/oai-reverse-proxy!54) 2023-11-19 05:06:21 +00:00
nai-degen 6c02e9b265 don't enqueue requests which fail stream check 2023-11-17 14:36:47 -06:00
nai-degen bfd7e23124 encodes queue payload 2023-11-16 01:19:01 -06:00
khanon 6aa6bebf08 Scale SSE heartbeat size with traffic (khanon/oai-reverse-proxy!53) 2023-11-16 05:45:35 +00:00
nai-degen 6acdf35914 removes length from stalled request error message 2023-11-15 17:18:51 -06:00
nai-degen 5fabe1d1f8 uses exponential moving average for wait time calculation 2023-11-14 01:36:11 -06:00
khanon 20c064394a OpenAI DALL-E Image Generation (khanon/oai-reverse-proxy!52) 2023-11-14 05:41:19 +00:00
nai-degen 0d5dfeccf8 adds gpt4-turbo model family and support for gpt-4-1106-preview model 2023-11-06 15:29:43 -06:00
nai-degen 89e1ed46d5 re-signs AWS requests on every attempt to fix fucked up queueing 2023-10-24 13:10:50 -05:00
nai-degen 725fd6e6f1 deprioritizes queued Agnai.chat requests and limits concurrency to five across all shared IPs 2023-10-09 12:36:54 -05:00
nai-degen daf6a123d5 adjusts Agnai.chat and RisuAI rate limiting 2023-10-04 09:39:59 -05:00
nai-degen 5033d00444 improves clarity of errors sent back to streaming clients 2023-10-03 19:45:15 -05:00
khanon ecf897e685 Refactor handleStreamingResponse to make it less shit (khanon/oai-reverse-proxy!46) 2023-10-03 06:14:19 +00:00
khanon fa4bf468d2 Implement AWS Bedrock support (khanon/oai-reverse-proxy!45) 2023-10-01 01:40:18 +00:00
khanon 35a6c393ed Add support for Google PaLM and OpenAI Turbo Instruct (khanon/oai-reverse-proxy!44) 2023-09-19 23:13:08 +00:00
nai-degen 5e57dbb8f1 attempts to improve compatibility with BetterGPT frontend 2023-09-16 11:04:40 -05:00
khanon f05e196994 Refactor project structure and add user self-serve UI (khanon/oai-reverse-proxy!41) 2023-09-02 19:36:44 +00:00
khanon 4d781e1720 Add GPT-4-32k support (khanon/oai-reverse-proxy!39) 2023-08-29 22:56:54 +00:00
nai-degen 6bb67281d9 removes QUEUE_MODE config (now always enabled) 2023-08-09 18:29:34 -05:00
nai-degen e2bd8a6b86 extracts Risu auth into new middleware so queue can use it too 2023-07-22 13:48:02 -05:00
khanon 4f2a12ef14 Show per-model queues and keys on info page (khanon/oai-reverse-proxy!22) 2023-06-08 18:50:04 +00:00
khanon dae1262f7a Refactor request middleware (khanon/oai-reverse-proxy!18) 2023-06-02 04:03:16 +00:00
khanon 6723cbf662 Anthropic endpoint improvements (khanon/oai-reverse-proxy!16) 2023-05-30 03:13:17 +00:00
khanon 2d93463247 Implement support for Anthropic keys and Claude API (khanon/oai-reverse-proxy!15) 2023-05-29 17:08:08 +00:00
nai-degen 03aaa6daad wraps SSE error responses in code block backticks 2023-05-23 12:28:06 -05:00
nai-degen 13b6a3d7b8 adds header to improve nginx compatibility 2023-05-23 11:57:14 -05:00
nai-degen 03616f4bbc increases wait time estimation window to 5min 2023-05-23 11:26:40 -05:00
nai-degen 26a6e4cadb increases wait time calculation window 2023-05-22 19:33:03 -05:00
nai-degen 2bad644772 Prefer user tokens as rate-limit/queue keys when available (khanon/oai-reverse-proxy!10) 2023-05-19 04:33:20 +00:00
nai-degen f1ac64fa12 Implement user persistence via Firebase (khanon/oai-reverse-proxy!8) 2023-05-14 04:26:08 +00:00
nai-degen 546b28cca6 increases wait time calc window to 90sec 2023-05-13 13:32:37 -05:00
nai-degen 09184079af reduces wait time window from 2min to 1min 2023-05-11 11:45:15 -05:00
nai-degen e03f3d48dd Implements request queueing (khanon/oai-reverse-proxy!6) 2023-05-09 23:11:57 +00:00