llama-stack

phoenix-oss/llama-stack

Fork 0

forked from phoenix-oss/llama-stack-mirror

Commit graph

56aed59eb4

Support for Llama3.2 models and Swift SDK (#98) Ashwin Bharambe 2024-09-25 10:29:58 -07:00
95abbf576b

Bump version to 0.0.24 (#94) poegej 2024-09-25 09:31:12 -07:00
ed8d10775a Remove key Ashwin Bharambe 2024-09-25 05:53:37 -07:00
45be9f3b85 fix agent's embedding model config Xi Yan 2024-09-24 22:49:30 -07:00
f45705cd10 Some lightweight cleanup and renaming for bedrock safety adapter Ashwin Bharambe 2024-09-24 19:27:03 -07:00
a2465f3f9c Revert parts of 0d2eb3bd25 Ashwin Bharambe 2024-09-24 19:20:26 -07:00
059e50b389

[aws-bedrock] Support for Bedrock Safety adapter (#96) rsgrewal-aws 2024-09-24 19:16:55 -07:00
b85d675c6f Adding safety adapter for Together Yogish Baliga 2024-09-20 09:35:01 -07:00
0d2eb3bd25 Use inference APIs for running llama guard Ashwin Bharambe 2024-09-24 17:02:57 -07:00
c4534217c8 fix cli describe Xi Yan 2024-09-24 14:41:13 -07:00
00352bd251 Respect passed in embedding model Ashwin Bharambe 2024-09-24 14:40:28 -07:00
bda974e660 Make the "all-remote" distribution lightweight in dependencies and size Ashwin Bharambe 2024-09-24 14:18:57 -07:00
445536de64 Add httpx to core server deps Ashwin Bharambe 2024-09-24 10:41:47 -07:00
7b35a4c827 Bump version to 0.0.24 Ashwin Bharambe 2024-09-24 10:15:20 -07:00
8d511cdf91 Make build_conda_env a bit more robust Ashwin Bharambe 2024-09-24 10:10:22 -07:00
cd850c16de Bump version to 0.0.23 Ashwin Bharambe 2024-09-24 09:08:40 -07:00
d04cd97aba remove providers/impls/sqlite/* Xi Yan 2024-09-24 01:03:40 -07:00
e617273d8c attribute changed (model_args -> arch_args) Ashwin Bharambe 2024-09-23 21:44:26 -07:00
f136f802b1 Somewhat better error handling Ashwin Bharambe 2024-09-23 21:39:47 -07:00
f92ff86b96 fix shields in agents safety Xi Yan 2024-09-23 21:22:22 -07:00
c9005e95ed Another attempt at a proper bugfix for safety violations Ashwin Bharambe 2024-09-23 19:06:30 -07:00
e5bdd6615a bug fix for safety violation Xi Yan 2024-09-23 18:17:15 -07:00
70fb70a71c fix URL issue with agents Xi Yan 2024-09-23 16:44:25 -07:00
9eb5ec3e4b Bump version to 0.0.21 Ashwin Bharambe 2024-09-23 14:23:21 -07:00
ec4fc800cc

[API Updates] Model / shield / memory-bank routing + agent persistence + support for private headers (#92) Ashwin Bharambe 2024-09-23 14:22:22 -07:00
8bf8c07eb3 Respect user sent instructions in agent config and add them to system prompt Hardik Shah 2024-09-21 16:46:10 -07:00
06abd7e6c8 update MemoryToolDefinition Xi Yan 2024-09-20 17:51:53 -07:00
942cb87a3c remove apis/stack.py Ashwin Bharambe 2024-09-20 09:37:08 -07:00
33db4d2e45 ignore config dir Hardik Shah 2024-09-20 00:24:49 -07:00
7e9e6117e3 do not assume CONDA_PREFIX exists during configuration Hardik Shah 2024-09-19 23:26:51 -07:00
8fa49593e0

Allow TGI adaptor to have non-standard llama model names (#84) Hardik Shah 2024-09-19 21:42:15 -07:00
42d29f3a5a Allow TGI adaptor to have non-standard llama model names Hardik Shah 2024-09-19 21:36:10 -07:00
59af1c8fec

fix memory url parsing (#81) Xi Yan 2024-09-19 13:35:03 -07:00
132f9429b1 Add a test for CLI, but not fully done so disabled Ashwin Bharambe 2024-09-19 13:27:07 -07:00
8b3ffa33de Add another test case Ashwin Bharambe 2024-09-19 13:02:57 -07:00
abb43936ab Add a test runner and 2 very simple tests for agents Ashwin Bharambe 2024-09-19 12:05:36 -07:00
543222ac39 update inference prompt msg Xi Yan 2024-09-19 12:03:18 -07:00
a30b919ae1 update inference prompt msg Xi Yan 2024-09-19 12:01:46 -07:00
9eb01dd664 Add DOCKER_BINARY / DOCKER_OPTS to all scripts Ashwin Bharambe 2024-09-19 10:26:41 -07:00
ca4b87aa05 fix memory client Xi Yan 2024-09-19 09:29:40 -07:00
6302a1ee90

fix prompt with name args (#80) Xi Yan 2024-09-18 23:48:31 -07:00
c63d6cbd08 list(...keys()) so dict_keys does not show up Ashwin Bharambe 2024-09-18 23:23:36 -07:00
880ed37026

Update cli_reference.md Xi Yan 2024-09-18 23:05:24 -07:00
5c4a2dc0e1

Update getting_started.md Xi Yan 2024-09-18 23:03:14 -07:00
f5eda1decf Add default for max_seq_len Ashwin Bharambe 2024-09-18 21:59:10 -07:00
9ab27e852b Bug fixes for memory Ashwin Bharambe 2024-09-18 21:54:02 -07:00
8cdc2f0cfb No RunShieldRequest Ashwin Bharambe 2024-09-18 20:38:21 -07:00
f3f5873e9e regenerate openapi spec Xi Yan 2024-09-18 19:28:05 -07:00
9f1be108ce Bump version to 0.0.20 Xi Yan 2024-09-18 19:06:07 -07:00
455a6e4bb9 update MANIFEST Xi Yan 2024-09-18 18:58:44 -07:00
dff9eab48f Remove "APIs to serve" prompt Ashwin Bharambe 2024-09-18 18:26:26 -07:00
f5d5e32d62 fix docker configure Xi Yan 2024-09-18 17:23:37 -07:00
5ec64ac68c moving rfc->docs Xi Yan 2024-09-18 16:54:24 -07:00
2c1ad10710 move openapi from rfcs->docs Xi Yan 2024-09-18 16:09:17 -07:00
21058be0c1 Bump version to 0.0.19 Xi Yan 2024-09-18 15:48:38 -07:00
45e20ff431 update getting started Xi Yan 2024-09-18 15:40:48 -07:00
2f9e952813 update getting started guide Xi Yan 2024-09-18 15:35:54 -07:00
29ce73ff7a update requirements, added prompt-toolkit Hardik Shah 2024-09-18 15:21:45 -07:00
1128f69674

CLI: add build templates support, move imports (#77) Xi Yan 2024-09-18 14:25:53 -07:00
6b21523c28

CLI - add back build wizard, configure with name instead of build.yaml (#74) Xi Yan 2024-09-18 11:41:56 -07:00
e6fdb9df29

fix context retriever (#75) Xi Yan 2024-09-18 08:24:36 -07:00
055770a791 Stop asking for "apis to serve" as part of configure Ashwin Bharambe 2024-09-17 22:40:51 -07:00
eea0a83bd1

Update getting_started.md Dalton Flanagan 2024-09-18 00:47:41 -04:00
9fd431e710 make shield imports more lazy Ashwin Bharambe 2024-09-17 21:27:16 -07:00
81ff7476d3 Bump version to 0.0.18 Ashwin Bharambe 2024-09-17 20:08:04 -07:00
3e27131a69 Don't import pkg_resources until you need it Ashwin Bharambe 2024-09-17 20:01:22 -07:00
25adc83de8 Fix for safety Ashwin Bharambe 2024-09-17 19:56:58 -07:00
9487ad8294

API Updates (#73) Ashwin Bharambe 2024-09-17 19:51:35 -07:00
f294eac5f5 Bump version to 0.0.17 Xi Yan 2024-09-16 13:10:05 -07:00
5839c61002 stage back models api Xi Yan 2024-09-16 13:00:39 -07:00
82b5c0460e models api Xi Yan 2024-09-12 11:29:19 -07:00
a36699cd11 Rename the "package" word away Ashwin Bharambe 2024-09-16 12:22:01 -07:00
98c55b63b4 delete distribution registry Xi Yan 2024-09-16 12:11:59 -07:00
6f5d9a3df8 provider_type -> provider_id ... less confusing Ashwin Bharambe 2024-09-16 12:09:53 -07:00
ce6c868499

Update cli_reference.md Xi Yan 2024-09-16 12:02:46 -07:00
ed4272e31e

Update getting_started.md Xi Yan 2024-09-16 11:55:10 -07:00
d9147f3184

CLI Update: build -> configure -> run (#69) Xi Yan 2024-09-16 11:02:26 -07:00
73b71d9689 Handle Annotated types more correctly Ashwin Bharambe 2024-09-14 12:22:13 -07:00
53ab18d6bb Bump version to 0.0.16 Ashwin Bharambe 2024-09-14 08:09:45 -07:00
49ce36426f Make llama model download error message a bit better Ashwin Bharambe 2024-09-14 08:06:34 -07:00
7a283ea076 Bump version to 0.0.15 Ashwin Bharambe 2024-09-13 17:23:12 -07:00
498cf03617 add pypdf Ashwin Bharambe 2024-09-13 17:04:43 -07:00
19a14cd273 Nuke hardware_requirements from SKUs Ashwin Bharambe 2024-09-13 16:39:02 -07:00
d8b3fdbd54

Update README.md raghotham 2024-09-13 08:56:47 -07:00
6a863f9b78 Bump version to 0.0.14 Xi Yan 2024-09-12 21:24:07 -07:00
16635508bd Bump version to 0.0.14 Xi Yan 2024-09-12 15:11:15 -07:00
5712566061

Remove request wrapper migration (#64) Xi Yan 2024-09-12 15:03:49 -07:00
1d0e91d802

Support data: in URL for memory. Add ootb support for pdfs (#67) Hardik Shah 2024-09-12 13:00:21 -07:00
736092f6bc

[Inference] Use huggingface_hub inference client for TGI adapter (#53) Celina Hanouti 2024-09-12 18:11:35 +02:00
191cd28831

Simplified Telemetry API and tying it to logger (#57) Ashwin Bharambe 2024-09-11 14:25:37 -07:00
1433aaf9f7 add CODEOWNERS file Xi Yan 2024-09-11 11:40:37 -07:00
89300df5dc

Add config file based CLI (#60) Xi Yan 2024-09-11 11:39:46 -07:00
58def874a9

add safety to openapi spec (#62) Xi Yan 2024-09-10 17:47:13 -07:00
a11d92601b

Enable Bing search (#59) Hardik Shah 2024-09-10 12:34:29 -07:00
2b63074676 add /inference/chat_completion to SSE special case Dalton Flanagan 2024-09-10 01:14:11 -04:00
4f021de10f

API spec update, client demo with Stainless SDK (#58) Xi Yan 2024-09-09 13:09:47 -07:00
741310f78e rename observability -> Telemetry; regen Spec Ashwin Bharambe 2024-09-07 15:23:53 -07:00
70e682fbdf Update distribution_id -> distribution_type, provider_id -> provider_type Ashwin Bharambe 2024-09-07 08:42:28 -07:00
3f090d1975

Add Chroma and PGVector adapters (#56) Ashwin Bharambe 2024-09-06 18:53:17 -07:00
5de6ed946e

Query generators for RAG query (#54) Hardik Shah 2024-09-06 13:10:39 -07:00