1755 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
mudler	f43aeeb4a1	Add both API endpoints (completion, chat)	2023-04-09 12:30:55 +02:00
mudler	c17dcc5e9d	Allow to inject prompt as part of the call	2023-04-09 09:36:19 +02:00
mudler	4a932483e1	Small fixup to template loading	2023-04-08 11:59:40 +02:00
mudler	b710147b95	Add mutex on same models (parallel isn't supported yet)	2023-04-08 11:45:36 +02:00
mudler	ba70363330	Use template input	2023-04-08 11:24:25 +02:00
mudler	9fb581739b	Allow to template model prompts inputs	2023-04-08 10:46:51 +02:00
mudler	48aca246e3	Drop unused interactive mode	2023-04-07 11:31:14 +02:00
mudler	12eee097b7	Make it compatible with openAI api, support multiple models ... Signed-off-by: mudler <mudler@c3os.io>	2023-04-07 11:30:59 +02:00
mudler	b33d015b8c	Use go-llama.cpp	2023-04-07 10:08:15 +02:00
Ettore Di Giacinto	b7c0a108f5	Update README.md	2023-04-05 22:28:03 +02:00
Ettore Di Giacinto	f694a89c28	Update README.md	2023-04-05 22:14:00 +02:00
Ettore Di Giacinto	be682e6c2f	Update README.md ... Add short-term roadmap and mention webui	2023-04-05 22:04:35 +02:00
mudler	bf85a31f9e	Don't set a default model path	2023-04-05 22:00:15 +02:00
Ettore Di Giacinto	d69048e0b0	Update README.md	2023-04-05 00:41:02 +02:00
mudler	827f189163	Update README	2023-03-30 18:46:11 +02:00
mudler	a23deb5ec7	Drop duplicate target	2023-03-29 19:44:41 +02:00
mudler	999676b106	Add gpt4all instructions	2023-03-29 18:58:54 +02:00
mudler	c61b023bc8	Drop fat images, will document how to consume models	2023-03-29 18:55:24 +02:00
mudler	650a22aef1	Add compatibility to gpt4all models	2023-03-29 18:53:24 +02:00
mudler	17b1724f7c	Update llama-go	2023-03-27 01:18:14 +02:00
mudler	e860e62036	Add mutex, build only lite images	2023-03-27 01:01:38 +02:00
Ettore Di Giacinto	1f45ff8cd6	Update README.md	2023-03-26 23:37:26 +02:00
mudler	abee34f60a	Cleanup leftover	2023-03-25 01:10:50 +01:00
mudler	dbc70dc13c	Add a simple web-page as index of the API for helping with inference testing	2023-03-25 01:09:51 +01:00
mudler	55142065eb	Update README with building instructions	2023-03-24 01:11:13 +01:00
mudler	d83d2293b5	Update version in kubernetes deployment	2023-03-23 23:22:43 +01:00
mudler	467ce5a7aa	Update models download instructions, update images	2023-03-23 22:06:41 +01:00
mudler	4c9c5ce4ce	Update README on instruction on how to prompt with the API	2023-03-23 19:25:28 +01:00
mudler	6394d85ca2	Lower conversion parallelism	2023-03-23 19:22:23 +01:00
mudler	2b6a5aef5f	Lower earthly parallelism	2023-03-23 19:17:15 +01:00
mudler	d191ecb9fe	Disable release pipeline	2023-03-23 19:14:39 +01:00
mudler	e14e1b0a77	Update README	2023-03-23 18:57:25 +01:00
mudler	bffaf2aa42	Build images without model	2023-03-23 18:50:43 +01:00
mudler	d98d1fe55e	Use models from model repository	2023-03-23 18:44:24 +01:00
mudler	0785cb6b0b	Update README with 13B and 30B model instructions	2023-03-22 00:18:48 +01:00
mudler	f88d5ad829	Update MODEL_URL	2023-03-21 22:03:20 +01:00
Ettore Di Giacinto	c7119a2882	Use tagged image in kubernetes deployment	2023-03-21 21:33:11 +01:00
mudler	8324402b49	Add interactive.go	2023-03-21 19:21:58 +01:00
mudler	9ba30c9c44	Update llama-go, allow to set context-size and enable alpaca model by default	2023-03-21 19:20:23 +01:00
mudler	973042bb4c	Update README to use tagged container images	2023-03-21 18:45:59 +01:00
mudler	3ed2888646	Update README	2023-03-20 23:26:29 +01:00
mudler	593ff6308c	Add simple client	2023-03-20 23:25:39 +01:00
mudler	4275bfc8c0	Add README	2023-03-20 21:30:55 +01:00
mudler	065815f947	Add kubernetes deployment sample	2023-03-20 21:30:38 +01:00
mudler	0460be964f	Fix entrypoint	2023-03-20 11:20:47 +01:00
mudler	6ca13f0227	Cleanup workers to have more free space	2023-03-20 10:12:31 +01:00
mudler	e6156b59fc	Cleanup	2023-03-20 00:46:49 +01:00
mudler	8da01d768c	Update Earthly versions	2023-03-20 00:40:32 +01:00
mudler	e764c3225c	Workaround Earthly issue	2023-03-20 00:24:37 +01:00
mudler	2ce1d51ad5	No need to set 0 for default context anymore	2023-03-20 00:12:26 +01:00