llama-stack 0.3.5__py3-none-any.whl → 0.4.0__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (458) hide show
  1. llama_stack/__init__.py +0 -5
  2. llama_stack/cli/llama.py +3 -3
  3. llama_stack/cli/stack/_list_deps.py +12 -23
  4. llama_stack/cli/stack/list_stacks.py +37 -18
  5. llama_stack/cli/stack/run.py +121 -11
  6. llama_stack/cli/stack/utils.py +0 -127
  7. llama_stack/core/access_control/access_control.py +69 -28
  8. llama_stack/core/access_control/conditions.py +15 -5
  9. llama_stack/core/admin.py +267 -0
  10. llama_stack/core/build.py +6 -74
  11. llama_stack/core/client.py +1 -1
  12. llama_stack/core/configure.py +6 -6
  13. llama_stack/core/conversations/conversations.py +28 -25
  14. llama_stack/core/datatypes.py +271 -79
  15. llama_stack/core/distribution.py +15 -16
  16. llama_stack/core/external.py +3 -3
  17. llama_stack/core/inspect.py +98 -15
  18. llama_stack/core/library_client.py +73 -61
  19. llama_stack/core/prompts/prompts.py +12 -11
  20. llama_stack/core/providers.py +17 -11
  21. llama_stack/core/resolver.py +65 -56
  22. llama_stack/core/routers/__init__.py +8 -12
  23. llama_stack/core/routers/datasets.py +1 -4
  24. llama_stack/core/routers/eval_scoring.py +7 -4
  25. llama_stack/core/routers/inference.py +55 -271
  26. llama_stack/core/routers/safety.py +52 -24
  27. llama_stack/core/routers/tool_runtime.py +6 -48
  28. llama_stack/core/routers/vector_io.py +130 -51
  29. llama_stack/core/routing_tables/benchmarks.py +24 -20
  30. llama_stack/core/routing_tables/common.py +1 -4
  31. llama_stack/core/routing_tables/datasets.py +22 -22
  32. llama_stack/core/routing_tables/models.py +119 -6
  33. llama_stack/core/routing_tables/scoring_functions.py +7 -7
  34. llama_stack/core/routing_tables/shields.py +1 -2
  35. llama_stack/core/routing_tables/toolgroups.py +17 -7
  36. llama_stack/core/routing_tables/vector_stores.py +51 -16
  37. llama_stack/core/server/auth.py +5 -3
  38. llama_stack/core/server/auth_providers.py +36 -20
  39. llama_stack/core/server/fastapi_router_registry.py +84 -0
  40. llama_stack/core/server/quota.py +2 -2
  41. llama_stack/core/server/routes.py +79 -27
  42. llama_stack/core/server/server.py +102 -87
  43. llama_stack/core/stack.py +201 -58
  44. llama_stack/core/storage/datatypes.py +26 -3
  45. llama_stack/{providers/utils → core/storage}/kvstore/__init__.py +2 -0
  46. llama_stack/{providers/utils → core/storage}/kvstore/kvstore.py +55 -24
  47. llama_stack/{providers/utils → core/storage}/kvstore/mongodb/mongodb.py +13 -10
  48. llama_stack/{providers/utils → core/storage}/kvstore/postgres/postgres.py +28 -17
  49. llama_stack/{providers/utils → core/storage}/kvstore/redis/redis.py +41 -16
  50. llama_stack/{providers/utils → core/storage}/kvstore/sqlite/sqlite.py +1 -1
  51. llama_stack/core/storage/sqlstore/__init__.py +17 -0
  52. llama_stack/{providers/utils → core/storage}/sqlstore/authorized_sqlstore.py +69 -49
  53. llama_stack/{providers/utils → core/storage}/sqlstore/sqlalchemy_sqlstore.py +47 -17
  54. llama_stack/{providers/utils → core/storage}/sqlstore/sqlstore.py +25 -8
  55. llama_stack/core/store/registry.py +1 -1
  56. llama_stack/core/utils/config.py +8 -2
  57. llama_stack/core/utils/config_resolution.py +32 -29
  58. llama_stack/core/utils/context.py +4 -10
  59. llama_stack/core/utils/exec.py +9 -0
  60. llama_stack/core/utils/type_inspection.py +45 -0
  61. llama_stack/distributions/dell/{run.yaml → config.yaml} +3 -2
  62. llama_stack/distributions/dell/dell.py +2 -2
  63. llama_stack/distributions/dell/run-with-safety.yaml +3 -2
  64. llama_stack/distributions/meta-reference-gpu/{run.yaml → config.yaml} +3 -2
  65. llama_stack/distributions/meta-reference-gpu/meta_reference.py +2 -2
  66. llama_stack/distributions/meta-reference-gpu/run-with-safety.yaml +3 -2
  67. llama_stack/distributions/nvidia/{run.yaml → config.yaml} +4 -4
  68. llama_stack/distributions/nvidia/nvidia.py +1 -1
  69. llama_stack/distributions/nvidia/run-with-safety.yaml +4 -4
  70. llama_stack/{apis/datasetio → distributions/oci}/__init__.py +1 -1
  71. llama_stack/distributions/oci/config.yaml +134 -0
  72. llama_stack/distributions/oci/oci.py +108 -0
  73. llama_stack/distributions/open-benchmark/{run.yaml → config.yaml} +5 -4
  74. llama_stack/distributions/open-benchmark/open_benchmark.py +2 -3
  75. llama_stack/distributions/postgres-demo/{run.yaml → config.yaml} +4 -3
  76. llama_stack/distributions/starter/{run.yaml → config.yaml} +64 -13
  77. llama_stack/distributions/starter/run-with-postgres-store.yaml +64 -13
  78. llama_stack/distributions/starter/starter.py +8 -5
  79. llama_stack/distributions/starter-gpu/{run.yaml → config.yaml} +64 -13
  80. llama_stack/distributions/starter-gpu/run-with-postgres-store.yaml +64 -13
  81. llama_stack/distributions/template.py +13 -69
  82. llama_stack/distributions/watsonx/{run.yaml → config.yaml} +4 -3
  83. llama_stack/distributions/watsonx/watsonx.py +1 -1
  84. llama_stack/log.py +28 -11
  85. llama_stack/models/llama/checkpoint.py +6 -6
  86. llama_stack/models/llama/hadamard_utils.py +2 -0
  87. llama_stack/models/llama/llama3/generation.py +3 -1
  88. llama_stack/models/llama/llama3/interface.py +2 -5
  89. llama_stack/models/llama/llama3/multimodal/encoder_utils.py +3 -3
  90. llama_stack/models/llama/llama3/multimodal/image_transform.py +6 -6
  91. llama_stack/models/llama/llama3/prompt_templates/system_prompts.py +1 -1
  92. llama_stack/models/llama/llama3/tool_utils.py +2 -1
  93. llama_stack/models/llama/llama4/prompt_templates/system_prompts.py +1 -1
  94. llama_stack/providers/inline/agents/meta_reference/__init__.py +3 -3
  95. llama_stack/providers/inline/agents/meta_reference/agents.py +44 -261
  96. llama_stack/providers/inline/agents/meta_reference/config.py +6 -1
  97. llama_stack/providers/inline/agents/meta_reference/responses/openai_responses.py +207 -57
  98. llama_stack/providers/inline/agents/meta_reference/responses/streaming.py +308 -47
  99. llama_stack/providers/inline/agents/meta_reference/responses/tool_executor.py +162 -96
  100. llama_stack/providers/inline/agents/meta_reference/responses/types.py +23 -8
  101. llama_stack/providers/inline/agents/meta_reference/responses/utils.py +201 -33
  102. llama_stack/providers/inline/agents/meta_reference/safety.py +8 -13
  103. llama_stack/providers/inline/batches/reference/__init__.py +2 -4
  104. llama_stack/providers/inline/batches/reference/batches.py +78 -60
  105. llama_stack/providers/inline/datasetio/localfs/datasetio.py +2 -5
  106. llama_stack/providers/inline/eval/meta_reference/eval.py +16 -61
  107. llama_stack/providers/inline/files/localfs/files.py +37 -28
  108. llama_stack/providers/inline/inference/meta_reference/config.py +2 -2
  109. llama_stack/providers/inline/inference/meta_reference/generators.py +50 -60
  110. llama_stack/providers/inline/inference/meta_reference/inference.py +403 -19
  111. llama_stack/providers/inline/inference/meta_reference/model_parallel.py +7 -26
  112. llama_stack/providers/inline/inference/meta_reference/parallel_utils.py +2 -12
  113. llama_stack/providers/inline/inference/sentence_transformers/sentence_transformers.py +10 -15
  114. llama_stack/providers/inline/post_training/common/validator.py +1 -5
  115. llama_stack/providers/inline/post_training/huggingface/post_training.py +8 -8
  116. llama_stack/providers/inline/post_training/huggingface/recipes/finetune_single_device.py +18 -10
  117. llama_stack/providers/inline/post_training/huggingface/recipes/finetune_single_device_dpo.py +12 -9
  118. llama_stack/providers/inline/post_training/huggingface/utils.py +27 -6
  119. llama_stack/providers/inline/post_training/torchtune/common/checkpointer.py +1 -1
  120. llama_stack/providers/inline/post_training/torchtune/common/utils.py +1 -1
  121. llama_stack/providers/inline/post_training/torchtune/datasets/format_adapter.py +1 -1
  122. llama_stack/providers/inline/post_training/torchtune/post_training.py +8 -8
  123. llama_stack/providers/inline/post_training/torchtune/recipes/lora_finetuning_single_device.py +16 -16
  124. llama_stack/providers/inline/safety/code_scanner/code_scanner.py +13 -9
  125. llama_stack/providers/inline/safety/llama_guard/llama_guard.py +18 -15
  126. llama_stack/providers/inline/safety/prompt_guard/prompt_guard.py +9 -9
  127. llama_stack/providers/inline/scoring/basic/scoring.py +6 -13
  128. llama_stack/providers/inline/scoring/basic/scoring_fn/docvqa_scoring_fn.py +1 -2
  129. llama_stack/providers/inline/scoring/basic/scoring_fn/equality_scoring_fn.py +1 -2
  130. llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/docvqa.py +2 -2
  131. llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/equality.py +2 -2
  132. llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/ifeval.py +2 -2
  133. llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/regex_parser_math_response.py +2 -2
  134. llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/regex_parser_multiple_choice_answer.py +2 -2
  135. llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/subset_of.py +2 -2
  136. llama_stack/providers/inline/scoring/basic/scoring_fn/ifeval_scoring_fn.py +1 -2
  137. llama_stack/providers/inline/scoring/basic/scoring_fn/regex_parser_math_response_scoring_fn.py +1 -2
  138. llama_stack/providers/inline/scoring/basic/scoring_fn/regex_parser_scoring_fn.py +1 -2
  139. llama_stack/providers/inline/scoring/basic/scoring_fn/subset_of_scoring_fn.py +1 -2
  140. llama_stack/providers/inline/scoring/braintrust/braintrust.py +12 -15
  141. llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/answer_correctness.py +2 -2
  142. llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/answer_relevancy.py +2 -2
  143. llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/answer_similarity.py +2 -2
  144. llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/context_entity_recall.py +2 -2
  145. llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/context_precision.py +2 -2
  146. llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/context_recall.py +2 -2
  147. llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/context_relevancy.py +2 -2
  148. llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/factuality.py +2 -2
  149. llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/faithfulness.py +2 -2
  150. llama_stack/providers/inline/scoring/llm_as_judge/scoring.py +7 -14
  151. llama_stack/providers/inline/scoring/llm_as_judge/scoring_fn/fn_defs/llm_as_judge_405b_simpleqa.py +2 -2
  152. llama_stack/providers/inline/scoring/llm_as_judge/scoring_fn/fn_defs/llm_as_judge_base.py +1 -2
  153. llama_stack/providers/inline/scoring/llm_as_judge/scoring_fn/llm_as_judge_scoring_fn.py +1 -3
  154. llama_stack/providers/inline/tool_runtime/rag/__init__.py +1 -1
  155. llama_stack/providers/inline/tool_runtime/rag/config.py +8 -1
  156. llama_stack/providers/inline/tool_runtime/rag/context_retriever.py +7 -6
  157. llama_stack/providers/inline/tool_runtime/rag/memory.py +64 -48
  158. llama_stack/providers/inline/vector_io/chroma/__init__.py +1 -1
  159. llama_stack/providers/inline/vector_io/chroma/config.py +1 -1
  160. llama_stack/providers/inline/vector_io/faiss/__init__.py +1 -1
  161. llama_stack/providers/inline/vector_io/faiss/config.py +1 -1
  162. llama_stack/providers/inline/vector_io/faiss/faiss.py +43 -28
  163. llama_stack/providers/inline/vector_io/milvus/__init__.py +1 -1
  164. llama_stack/providers/inline/vector_io/milvus/config.py +1 -1
  165. llama_stack/providers/inline/vector_io/qdrant/__init__.py +1 -1
  166. llama_stack/providers/inline/vector_io/qdrant/config.py +1 -1
  167. llama_stack/providers/inline/vector_io/sqlite_vec/__init__.py +1 -1
  168. llama_stack/providers/inline/vector_io/sqlite_vec/sqlite_vec.py +40 -33
  169. llama_stack/providers/registry/agents.py +7 -3
  170. llama_stack/providers/registry/batches.py +1 -1
  171. llama_stack/providers/registry/datasetio.py +1 -1
  172. llama_stack/providers/registry/eval.py +1 -1
  173. llama_stack/{apis/datasets/__init__.py → providers/registry/file_processors.py} +5 -1
  174. llama_stack/providers/registry/files.py +11 -2
  175. llama_stack/providers/registry/inference.py +22 -3
  176. llama_stack/providers/registry/post_training.py +1 -1
  177. llama_stack/providers/registry/safety.py +1 -1
  178. llama_stack/providers/registry/scoring.py +1 -1
  179. llama_stack/providers/registry/tool_runtime.py +2 -2
  180. llama_stack/providers/registry/vector_io.py +7 -7
  181. llama_stack/providers/remote/datasetio/huggingface/huggingface.py +2 -5
  182. llama_stack/providers/remote/datasetio/nvidia/datasetio.py +1 -4
  183. llama_stack/providers/remote/eval/nvidia/eval.py +15 -9
  184. llama_stack/providers/remote/files/openai/__init__.py +19 -0
  185. llama_stack/providers/remote/files/openai/config.py +28 -0
  186. llama_stack/providers/remote/files/openai/files.py +253 -0
  187. llama_stack/providers/remote/files/s3/files.py +52 -30
  188. llama_stack/providers/remote/inference/anthropic/anthropic.py +2 -1
  189. llama_stack/providers/remote/inference/anthropic/config.py +1 -1
  190. llama_stack/providers/remote/inference/azure/azure.py +1 -3
  191. llama_stack/providers/remote/inference/azure/config.py +8 -7
  192. llama_stack/providers/remote/inference/bedrock/__init__.py +1 -1
  193. llama_stack/providers/remote/inference/bedrock/bedrock.py +82 -105
  194. llama_stack/providers/remote/inference/bedrock/config.py +24 -3
  195. llama_stack/providers/remote/inference/cerebras/cerebras.py +5 -5
  196. llama_stack/providers/remote/inference/cerebras/config.py +12 -5
  197. llama_stack/providers/remote/inference/databricks/config.py +13 -6
  198. llama_stack/providers/remote/inference/databricks/databricks.py +16 -6
  199. llama_stack/providers/remote/inference/fireworks/config.py +5 -5
  200. llama_stack/providers/remote/inference/fireworks/fireworks.py +1 -1
  201. llama_stack/providers/remote/inference/gemini/config.py +1 -1
  202. llama_stack/providers/remote/inference/gemini/gemini.py +13 -14
  203. llama_stack/providers/remote/inference/groq/config.py +5 -5
  204. llama_stack/providers/remote/inference/groq/groq.py +1 -1
  205. llama_stack/providers/remote/inference/llama_openai_compat/config.py +5 -5
  206. llama_stack/providers/remote/inference/llama_openai_compat/llama.py +8 -6
  207. llama_stack/providers/remote/inference/nvidia/__init__.py +1 -1
  208. llama_stack/providers/remote/inference/nvidia/config.py +21 -11
  209. llama_stack/providers/remote/inference/nvidia/nvidia.py +115 -3
  210. llama_stack/providers/remote/inference/nvidia/utils.py +1 -1
  211. llama_stack/providers/remote/inference/oci/__init__.py +17 -0
  212. llama_stack/providers/remote/inference/oci/auth.py +79 -0
  213. llama_stack/providers/remote/inference/oci/config.py +75 -0
  214. llama_stack/providers/remote/inference/oci/oci.py +162 -0
  215. llama_stack/providers/remote/inference/ollama/config.py +7 -5
  216. llama_stack/providers/remote/inference/ollama/ollama.py +17 -8
  217. llama_stack/providers/remote/inference/openai/config.py +4 -4
  218. llama_stack/providers/remote/inference/openai/openai.py +1 -1
  219. llama_stack/providers/remote/inference/passthrough/__init__.py +2 -2
  220. llama_stack/providers/remote/inference/passthrough/config.py +5 -10
  221. llama_stack/providers/remote/inference/passthrough/passthrough.py +97 -75
  222. llama_stack/providers/remote/inference/runpod/config.py +12 -5
  223. llama_stack/providers/remote/inference/runpod/runpod.py +2 -20
  224. llama_stack/providers/remote/inference/sambanova/config.py +5 -5
  225. llama_stack/providers/remote/inference/sambanova/sambanova.py +1 -1
  226. llama_stack/providers/remote/inference/tgi/config.py +7 -6
  227. llama_stack/providers/remote/inference/tgi/tgi.py +19 -11
  228. llama_stack/providers/remote/inference/together/config.py +5 -5
  229. llama_stack/providers/remote/inference/together/together.py +15 -12
  230. llama_stack/providers/remote/inference/vertexai/config.py +1 -1
  231. llama_stack/providers/remote/inference/vllm/config.py +5 -5
  232. llama_stack/providers/remote/inference/vllm/vllm.py +13 -14
  233. llama_stack/providers/remote/inference/watsonx/config.py +4 -4
  234. llama_stack/providers/remote/inference/watsonx/watsonx.py +21 -94
  235. llama_stack/providers/remote/post_training/nvidia/post_training.py +4 -4
  236. llama_stack/providers/remote/post_training/nvidia/utils.py +1 -1
  237. llama_stack/providers/remote/safety/bedrock/bedrock.py +6 -6
  238. llama_stack/providers/remote/safety/bedrock/config.py +1 -1
  239. llama_stack/providers/remote/safety/nvidia/config.py +1 -1
  240. llama_stack/providers/remote/safety/nvidia/nvidia.py +11 -5
  241. llama_stack/providers/remote/safety/sambanova/config.py +1 -1
  242. llama_stack/providers/remote/safety/sambanova/sambanova.py +6 -6
  243. llama_stack/providers/remote/tool_runtime/bing_search/bing_search.py +11 -6
  244. llama_stack/providers/remote/tool_runtime/brave_search/brave_search.py +12 -7
  245. llama_stack/providers/remote/tool_runtime/model_context_protocol/config.py +8 -2
  246. llama_stack/providers/remote/tool_runtime/model_context_protocol/model_context_protocol.py +57 -15
  247. llama_stack/providers/remote/tool_runtime/tavily_search/tavily_search.py +11 -6
  248. llama_stack/providers/remote/tool_runtime/wolfram_alpha/wolfram_alpha.py +11 -6
  249. llama_stack/providers/remote/vector_io/chroma/__init__.py +1 -1
  250. llama_stack/providers/remote/vector_io/chroma/chroma.py +125 -20
  251. llama_stack/providers/remote/vector_io/chroma/config.py +1 -1
  252. llama_stack/providers/remote/vector_io/milvus/__init__.py +1 -1
  253. llama_stack/providers/remote/vector_io/milvus/config.py +1 -1
  254. llama_stack/providers/remote/vector_io/milvus/milvus.py +27 -21
  255. llama_stack/providers/remote/vector_io/pgvector/__init__.py +1 -1
  256. llama_stack/providers/remote/vector_io/pgvector/config.py +1 -1
  257. llama_stack/providers/remote/vector_io/pgvector/pgvector.py +26 -18
  258. llama_stack/providers/remote/vector_io/qdrant/__init__.py +1 -1
  259. llama_stack/providers/remote/vector_io/qdrant/config.py +1 -1
  260. llama_stack/providers/remote/vector_io/qdrant/qdrant.py +141 -24
  261. llama_stack/providers/remote/vector_io/weaviate/__init__.py +1 -1
  262. llama_stack/providers/remote/vector_io/weaviate/config.py +1 -1
  263. llama_stack/providers/remote/vector_io/weaviate/weaviate.py +26 -21
  264. llama_stack/providers/utils/common/data_schema_validator.py +1 -5
  265. llama_stack/providers/utils/files/form_data.py +1 -1
  266. llama_stack/providers/utils/inference/embedding_mixin.py +1 -1
  267. llama_stack/providers/utils/inference/inference_store.py +7 -8
  268. llama_stack/providers/utils/inference/litellm_openai_mixin.py +79 -79
  269. llama_stack/providers/utils/inference/model_registry.py +1 -3
  270. llama_stack/providers/utils/inference/openai_compat.py +44 -1171
  271. llama_stack/providers/utils/inference/openai_mixin.py +68 -42
  272. llama_stack/providers/utils/inference/prompt_adapter.py +50 -265
  273. llama_stack/providers/utils/inference/stream_utils.py +23 -0
  274. llama_stack/providers/utils/memory/__init__.py +2 -0
  275. llama_stack/providers/utils/memory/file_utils.py +1 -1
  276. llama_stack/providers/utils/memory/openai_vector_store_mixin.py +181 -84
  277. llama_stack/providers/utils/memory/vector_store.py +39 -38
  278. llama_stack/providers/utils/pagination.py +1 -1
  279. llama_stack/providers/utils/responses/responses_store.py +15 -25
  280. llama_stack/providers/utils/scoring/aggregation_utils.py +1 -2
  281. llama_stack/providers/utils/scoring/base_scoring_fn.py +1 -2
  282. llama_stack/providers/utils/tools/mcp.py +93 -11
  283. llama_stack/telemetry/constants.py +27 -0
  284. llama_stack/telemetry/helpers.py +43 -0
  285. llama_stack/testing/api_recorder.py +25 -16
  286. {llama_stack-0.3.5.dist-info → llama_stack-0.4.0.dist-info}/METADATA +56 -54
  287. llama_stack-0.4.0.dist-info/RECORD +588 -0
  288. llama_stack-0.4.0.dist-info/top_level.txt +2 -0
  289. llama_stack_api/__init__.py +945 -0
  290. llama_stack_api/admin/__init__.py +45 -0
  291. llama_stack_api/admin/api.py +72 -0
  292. llama_stack_api/admin/fastapi_routes.py +117 -0
  293. llama_stack_api/admin/models.py +113 -0
  294. llama_stack_api/agents.py +173 -0
  295. llama_stack_api/batches/__init__.py +40 -0
  296. llama_stack_api/batches/api.py +53 -0
  297. llama_stack_api/batches/fastapi_routes.py +113 -0
  298. llama_stack_api/batches/models.py +78 -0
  299. llama_stack_api/benchmarks/__init__.py +43 -0
  300. llama_stack_api/benchmarks/api.py +39 -0
  301. llama_stack_api/benchmarks/fastapi_routes.py +109 -0
  302. llama_stack_api/benchmarks/models.py +109 -0
  303. {llama_stack/apis → llama_stack_api}/common/content_types.py +1 -43
  304. {llama_stack/apis → llama_stack_api}/common/errors.py +0 -8
  305. {llama_stack/apis → llama_stack_api}/common/job_types.py +1 -1
  306. llama_stack_api/common/responses.py +77 -0
  307. {llama_stack/apis → llama_stack_api}/common/training_types.py +1 -1
  308. {llama_stack/apis → llama_stack_api}/common/type_system.py +2 -14
  309. llama_stack_api/connectors.py +146 -0
  310. {llama_stack/apis/conversations → llama_stack_api}/conversations.py +23 -39
  311. {llama_stack/apis/datasetio → llama_stack_api}/datasetio.py +4 -8
  312. llama_stack_api/datasets/__init__.py +61 -0
  313. llama_stack_api/datasets/api.py +35 -0
  314. llama_stack_api/datasets/fastapi_routes.py +104 -0
  315. llama_stack_api/datasets/models.py +152 -0
  316. {llama_stack/providers → llama_stack_api}/datatypes.py +166 -10
  317. {llama_stack/apis/eval → llama_stack_api}/eval.py +8 -40
  318. llama_stack_api/file_processors/__init__.py +27 -0
  319. llama_stack_api/file_processors/api.py +64 -0
  320. llama_stack_api/file_processors/fastapi_routes.py +78 -0
  321. llama_stack_api/file_processors/models.py +42 -0
  322. llama_stack_api/files/__init__.py +35 -0
  323. llama_stack_api/files/api.py +51 -0
  324. llama_stack_api/files/fastapi_routes.py +124 -0
  325. llama_stack_api/files/models.py +107 -0
  326. {llama_stack/apis/inference → llama_stack_api}/inference.py +90 -194
  327. llama_stack_api/inspect_api/__init__.py +37 -0
  328. llama_stack_api/inspect_api/api.py +25 -0
  329. llama_stack_api/inspect_api/fastapi_routes.py +76 -0
  330. llama_stack_api/inspect_api/models.py +28 -0
  331. {llama_stack/apis/agents → llama_stack_api/internal}/__init__.py +3 -1
  332. llama_stack/providers/utils/kvstore/api.py → llama_stack_api/internal/kvstore.py +5 -0
  333. llama_stack_api/internal/sqlstore.py +79 -0
  334. {llama_stack/apis/models → llama_stack_api}/models.py +11 -9
  335. {llama_stack/apis/agents → llama_stack_api}/openai_responses.py +184 -27
  336. {llama_stack/apis/post_training → llama_stack_api}/post_training.py +7 -11
  337. {llama_stack/apis/prompts → llama_stack_api}/prompts.py +3 -4
  338. llama_stack_api/providers/__init__.py +33 -0
  339. llama_stack_api/providers/api.py +16 -0
  340. llama_stack_api/providers/fastapi_routes.py +57 -0
  341. llama_stack_api/providers/models.py +24 -0
  342. {llama_stack/apis/tools → llama_stack_api}/rag_tool.py +2 -52
  343. {llama_stack/apis → llama_stack_api}/resource.py +1 -1
  344. llama_stack_api/router_utils.py +160 -0
  345. {llama_stack/apis/safety → llama_stack_api}/safety.py +6 -9
  346. {llama_stack → llama_stack_api}/schema_utils.py +94 -4
  347. {llama_stack/apis/scoring → llama_stack_api}/scoring.py +3 -3
  348. {llama_stack/apis/scoring_functions → llama_stack_api}/scoring_functions.py +9 -6
  349. {llama_stack/apis/shields → llama_stack_api}/shields.py +6 -7
  350. {llama_stack/apis/tools → llama_stack_api}/tools.py +26 -21
  351. {llama_stack/apis/vector_io → llama_stack_api}/vector_io.py +133 -152
  352. {llama_stack/apis/vector_stores → llama_stack_api}/vector_stores.py +1 -1
  353. llama_stack/apis/agents/agents.py +0 -894
  354. llama_stack/apis/batches/__init__.py +0 -9
  355. llama_stack/apis/batches/batches.py +0 -100
  356. llama_stack/apis/benchmarks/__init__.py +0 -7
  357. llama_stack/apis/benchmarks/benchmarks.py +0 -108
  358. llama_stack/apis/common/responses.py +0 -36
  359. llama_stack/apis/conversations/__init__.py +0 -31
  360. llama_stack/apis/datasets/datasets.py +0 -251
  361. llama_stack/apis/datatypes.py +0 -160
  362. llama_stack/apis/eval/__init__.py +0 -7
  363. llama_stack/apis/files/__init__.py +0 -7
  364. llama_stack/apis/files/files.py +0 -199
  365. llama_stack/apis/inference/__init__.py +0 -7
  366. llama_stack/apis/inference/event_logger.py +0 -43
  367. llama_stack/apis/inspect/__init__.py +0 -7
  368. llama_stack/apis/inspect/inspect.py +0 -94
  369. llama_stack/apis/models/__init__.py +0 -7
  370. llama_stack/apis/post_training/__init__.py +0 -7
  371. llama_stack/apis/prompts/__init__.py +0 -9
  372. llama_stack/apis/providers/__init__.py +0 -7
  373. llama_stack/apis/providers/providers.py +0 -69
  374. llama_stack/apis/safety/__init__.py +0 -7
  375. llama_stack/apis/scoring/__init__.py +0 -7
  376. llama_stack/apis/scoring_functions/__init__.py +0 -7
  377. llama_stack/apis/shields/__init__.py +0 -7
  378. llama_stack/apis/synthetic_data_generation/__init__.py +0 -7
  379. llama_stack/apis/synthetic_data_generation/synthetic_data_generation.py +0 -77
  380. llama_stack/apis/telemetry/__init__.py +0 -7
  381. llama_stack/apis/telemetry/telemetry.py +0 -423
  382. llama_stack/apis/tools/__init__.py +0 -8
  383. llama_stack/apis/vector_io/__init__.py +0 -7
  384. llama_stack/apis/vector_stores/__init__.py +0 -7
  385. llama_stack/core/server/tracing.py +0 -80
  386. llama_stack/core/ui/app.py +0 -55
  387. llama_stack/core/ui/modules/__init__.py +0 -5
  388. llama_stack/core/ui/modules/api.py +0 -32
  389. llama_stack/core/ui/modules/utils.py +0 -42
  390. llama_stack/core/ui/page/__init__.py +0 -5
  391. llama_stack/core/ui/page/distribution/__init__.py +0 -5
  392. llama_stack/core/ui/page/distribution/datasets.py +0 -18
  393. llama_stack/core/ui/page/distribution/eval_tasks.py +0 -20
  394. llama_stack/core/ui/page/distribution/models.py +0 -18
  395. llama_stack/core/ui/page/distribution/providers.py +0 -27
  396. llama_stack/core/ui/page/distribution/resources.py +0 -48
  397. llama_stack/core/ui/page/distribution/scoring_functions.py +0 -18
  398. llama_stack/core/ui/page/distribution/shields.py +0 -19
  399. llama_stack/core/ui/page/evaluations/__init__.py +0 -5
  400. llama_stack/core/ui/page/evaluations/app_eval.py +0 -143
  401. llama_stack/core/ui/page/evaluations/native_eval.py +0 -253
  402. llama_stack/core/ui/page/playground/__init__.py +0 -5
  403. llama_stack/core/ui/page/playground/chat.py +0 -130
  404. llama_stack/core/ui/page/playground/tools.py +0 -352
  405. llama_stack/distributions/dell/build.yaml +0 -33
  406. llama_stack/distributions/meta-reference-gpu/build.yaml +0 -32
  407. llama_stack/distributions/nvidia/build.yaml +0 -29
  408. llama_stack/distributions/open-benchmark/build.yaml +0 -36
  409. llama_stack/distributions/postgres-demo/__init__.py +0 -7
  410. llama_stack/distributions/postgres-demo/build.yaml +0 -23
  411. llama_stack/distributions/postgres-demo/postgres_demo.py +0 -125
  412. llama_stack/distributions/starter/build.yaml +0 -61
  413. llama_stack/distributions/starter-gpu/build.yaml +0 -61
  414. llama_stack/distributions/watsonx/build.yaml +0 -33
  415. llama_stack/providers/inline/agents/meta_reference/agent_instance.py +0 -1024
  416. llama_stack/providers/inline/agents/meta_reference/persistence.py +0 -228
  417. llama_stack/providers/inline/telemetry/__init__.py +0 -5
  418. llama_stack/providers/inline/telemetry/meta_reference/__init__.py +0 -21
  419. llama_stack/providers/inline/telemetry/meta_reference/config.py +0 -47
  420. llama_stack/providers/inline/telemetry/meta_reference/telemetry.py +0 -252
  421. llama_stack/providers/remote/inference/bedrock/models.py +0 -29
  422. llama_stack/providers/utils/kvstore/sqlite/config.py +0 -20
  423. llama_stack/providers/utils/sqlstore/__init__.py +0 -5
  424. llama_stack/providers/utils/sqlstore/api.py +0 -128
  425. llama_stack/providers/utils/telemetry/__init__.py +0 -5
  426. llama_stack/providers/utils/telemetry/trace_protocol.py +0 -142
  427. llama_stack/providers/utils/telemetry/tracing.py +0 -384
  428. llama_stack/strong_typing/__init__.py +0 -19
  429. llama_stack/strong_typing/auxiliary.py +0 -228
  430. llama_stack/strong_typing/classdef.py +0 -440
  431. llama_stack/strong_typing/core.py +0 -46
  432. llama_stack/strong_typing/deserializer.py +0 -877
  433. llama_stack/strong_typing/docstring.py +0 -409
  434. llama_stack/strong_typing/exception.py +0 -23
  435. llama_stack/strong_typing/inspection.py +0 -1085
  436. llama_stack/strong_typing/mapping.py +0 -40
  437. llama_stack/strong_typing/name.py +0 -182
  438. llama_stack/strong_typing/schema.py +0 -792
  439. llama_stack/strong_typing/serialization.py +0 -97
  440. llama_stack/strong_typing/serializer.py +0 -500
  441. llama_stack/strong_typing/slots.py +0 -27
  442. llama_stack/strong_typing/topological.py +0 -89
  443. llama_stack/ui/node_modules/flatted/python/flatted.py +0 -149
  444. llama_stack-0.3.5.dist-info/RECORD +0 -625
  445. llama_stack-0.3.5.dist-info/top_level.txt +0 -1
  446. /llama_stack/{providers/utils → core/storage}/kvstore/config.py +0 -0
  447. /llama_stack/{providers/utils → core/storage}/kvstore/mongodb/__init__.py +0 -0
  448. /llama_stack/{providers/utils → core/storage}/kvstore/postgres/__init__.py +0 -0
  449. /llama_stack/{providers/utils → core/storage}/kvstore/redis/__init__.py +0 -0
  450. /llama_stack/{providers/utils → core/storage}/kvstore/sqlite/__init__.py +0 -0
  451. /llama_stack/{apis → providers/inline/file_processor}/__init__.py +0 -0
  452. /llama_stack/{apis/common → telemetry}/__init__.py +0 -0
  453. {llama_stack-0.3.5.dist-info → llama_stack-0.4.0.dist-info}/WHEEL +0 -0
  454. {llama_stack-0.3.5.dist-info → llama_stack-0.4.0.dist-info}/entry_points.txt +0 -0
  455. {llama_stack-0.3.5.dist-info → llama_stack-0.4.0.dist-info}/licenses/LICENSE +0 -0
  456. {llama_stack/core/ui → llama_stack_api/common}/__init__.py +0 -0
  457. {llama_stack/strong_typing → llama_stack_api}/py.typed +0 -0
  458. {llama_stack/apis → llama_stack_api}/version.py +0 -0
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,9 +4,9 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
9
+ NumberType,
10
10
  RegexParserScoringFnParams,
11
11
  ScoringFn,
12
12
  )
@@ -4,9 +4,9 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
9
+ NumberType,
10
10
  RegexParserScoringFnParams,
11
11
  ScoringFn,
12
12
  )
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -6,9 +6,8 @@
6
6
 
7
7
  from typing import Any
8
8
 
9
- from llama_stack.apis.scoring import ScoringResultRow
10
- from llama_stack.apis.scoring_functions import ScoringFnParams
11
9
  from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
10
+ from llama_stack_api import ScoringFnParams, ScoringResultRow
12
11
 
13
12
  from .fn_defs.ifeval import (
14
13
  ifeval,
@@ -5,9 +5,8 @@
5
5
  # the root directory of this source tree.
6
6
  from typing import Any
7
7
 
8
- from llama_stack.apis.scoring import ScoringResultRow
9
- from llama_stack.apis.scoring_functions import ScoringFnParams, ScoringFnParamsType
10
8
  from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
9
+ from llama_stack_api import ScoringFnParams, ScoringFnParamsType, ScoringResultRow
11
10
 
12
11
  from ..utils.math_utils import first_answer, normalize_final_answer, try_evaluate_frac, try_evaluate_latex
13
12
  from .fn_defs.regex_parser_math_response import (
@@ -6,9 +6,8 @@
6
6
  import re
7
7
  from typing import Any
8
8
 
9
- from llama_stack.apis.scoring import ScoringResultRow
10
- from llama_stack.apis.scoring_functions import ScoringFnParams, ScoringFnParamsType
11
9
  from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
10
+ from llama_stack_api import ScoringFnParams, ScoringFnParamsType, ScoringResultRow
12
11
 
13
12
  from .fn_defs.regex_parser_multiple_choice_answer import (
14
13
  regex_parser_multiple_choice_answer,
@@ -6,9 +6,8 @@
6
6
 
7
7
  from typing import Any
8
8
 
9
- from llama_stack.apis.scoring import ScoringResultRow
10
- from llama_stack.apis.scoring_functions import ScoringFnParams
11
9
  from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
10
+ from llama_stack_api import ScoringFnParams, ScoringResultRow
12
11
 
13
12
  from .fn_defs.subset_of import subset_of
14
13
 
@@ -19,25 +19,25 @@ from autoevals.ragas import (
19
19
  )
20
20
  from pydantic import BaseModel
21
21
 
22
- from llama_stack.apis.datasetio import DatasetIO
23
- from llama_stack.apis.datasets import Datasets
24
- from llama_stack.apis.scoring import (
25
- ScoreBatchResponse,
26
- ScoreResponse,
27
- Scoring,
28
- ScoringResult,
29
- ScoringResultRow,
30
- )
31
- from llama_stack.apis.scoring_functions import ScoringFn, ScoringFnParams
32
22
  from llama_stack.core.datatypes import Api
33
23
  from llama_stack.core.request_headers import NeedsRequestProviderData
34
- from llama_stack.providers.datatypes import ScoringFunctionsProtocolPrivate
35
24
  from llama_stack.providers.utils.common.data_schema_validator import (
36
25
  get_valid_schemas,
37
- validate_dataset_schema,
38
26
  validate_row_schema,
39
27
  )
40
28
  from llama_stack.providers.utils.scoring.aggregation_utils import aggregate_metrics
29
+ from llama_stack_api import (
30
+ DatasetIO,
31
+ Datasets,
32
+ ScoreBatchResponse,
33
+ ScoreResponse,
34
+ Scoring,
35
+ ScoringFn,
36
+ ScoringFnParams,
37
+ ScoringFunctionsProtocolPrivate,
38
+ ScoringResult,
39
+ ScoringResultRow,
40
+ )
41
41
 
42
42
  from .config import BraintrustScoringConfig
43
43
  from .scoring_fn.fn_defs.answer_correctness import answer_correctness_fn_def
@@ -164,9 +164,6 @@ class BraintrustScoringImpl(
164
164
  ) -> ScoreBatchResponse:
165
165
  await self.set_api_key()
166
166
 
167
- dataset_def = await self.datasets_api.get_dataset(dataset_id=dataset_id)
168
- validate_dataset_schema(dataset_def.dataset_schema, get_valid_schemas(Api.scoring.value))
169
-
170
167
  all_rows = await self.datasetio_api.iterrows(
171
168
  dataset_id=dataset_id,
172
169
  limit=-1,
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  BasicScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -5,22 +5,18 @@
5
5
  # the root directory of this source tree.
6
6
  from typing import Any
7
7
 
8
- from llama_stack.apis.datasetio import DatasetIO
9
- from llama_stack.apis.datasets import Datasets
10
- from llama_stack.apis.inference import Inference
11
- from llama_stack.apis.scoring import (
8
+ from llama_stack_api import (
9
+ DatasetIO,
10
+ Datasets,
11
+ Inference,
12
12
  ScoreBatchResponse,
13
13
  ScoreResponse,
14
14
  Scoring,
15
+ ScoringFn,
16
+ ScoringFnParams,
17
+ ScoringFunctionsProtocolPrivate,
15
18
  ScoringResult,
16
19
  )
17
- from llama_stack.apis.scoring_functions import ScoringFn, ScoringFnParams
18
- from llama_stack.core.datatypes import Api
19
- from llama_stack.providers.datatypes import ScoringFunctionsProtocolPrivate
20
- from llama_stack.providers.utils.common.data_schema_validator import (
21
- get_valid_schemas,
22
- validate_dataset_schema,
23
- )
24
20
 
25
21
  from .config import LlmAsJudgeScoringConfig
26
22
  from .scoring_fn.llm_as_judge_scoring_fn import LlmAsJudgeScoringFn
@@ -72,9 +68,6 @@ class LlmAsJudgeScoringImpl(
72
68
  scoring_functions: dict[str, ScoringFnParams | None] = None,
73
69
  save_results_dataset: bool = False,
74
70
  ) -> ScoreBatchResponse:
75
- dataset_def = await self.datasets_api.get_dataset(dataset_id=dataset_id)
76
- validate_dataset_schema(dataset_def.dataset_schema, get_valid_schemas(Api.scoring.value))
77
-
78
71
  all_rows = await self.datasetio_api.iterrows(
79
72
  dataset_id=dataset_id,
80
73
  limit=-1,
@@ -4,10 +4,10 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import (
7
+ from llama_stack_api import (
9
8
  AggregationFunctionType,
10
9
  LLMAsJudgeScoringFnParams,
10
+ NumberType,
11
11
  ScoringFn,
12
12
  )
13
13
 
@@ -4,8 +4,7 @@
4
4
  # This source code is licensed under the terms described in the LICENSE file in
5
5
  # the root directory of this source tree.
6
6
 
7
- from llama_stack.apis.common.type_system import NumberType
8
- from llama_stack.apis.scoring_functions import LLMAsJudgeScoringFnParams, ScoringFn
7
+ from llama_stack_api import LLMAsJudgeScoringFnParams, NumberType, ScoringFn
9
8
 
10
9
  llm_as_judge_base = ScoringFn(
11
10
  identifier="llm-as-judge::base",
@@ -6,10 +6,8 @@
6
6
  import re
7
7
  from typing import Any
8
8
 
9
- from llama_stack.apis.inference import Inference, OpenAIChatCompletionRequestWithExtraBody
10
- from llama_stack.apis.scoring import ScoringResultRow
11
- from llama_stack.apis.scoring_functions import ScoringFnParams
12
9
  from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
10
+ from llama_stack_api import Inference, OpenAIChatCompletionRequestWithExtraBody, ScoringFnParams, ScoringResultRow
13
11
 
14
12
  from .fn_defs.llm_as_judge_405b_simpleqa import llm_as_judge_405b_simpleqa
15
13
  from .fn_defs.llm_as_judge_base import llm_as_judge_base
@@ -6,7 +6,7 @@
6
6
 
7
7
  from typing import Any
8
8
 
9
- from llama_stack.providers.datatypes import Api
9
+ from llama_stack_api import Api
10
10
 
11
11
  from .config import RagToolRuntimeConfig
12
12
 
@@ -6,10 +6,17 @@
6
6
 
7
7
  from typing import Any
8
8
 
9
- from pydantic import BaseModel
9
+ from pydantic import BaseModel, Field
10
+
11
+ from llama_stack.core.datatypes import VectorStoresConfig
10
12
 
11
13
 
12
14
  class RagToolRuntimeConfig(BaseModel):
15
+ vector_stores_config: VectorStoresConfig = Field(
16
+ default_factory=VectorStoresConfig,
17
+ description="Configuration for vector store prompt templates and behavior",
18
+ )
19
+
13
20
  @classmethod
14
21
  def sample_run_config(cls, __distro_dir__: str, **kwargs: Any) -> dict[str, Any]:
15
22
  return {}
@@ -7,17 +7,18 @@
7
7
 
8
8
  from jinja2 import Template
9
9
 
10
- from llama_stack.apis.common.content_types import InterleavedContent
11
- from llama_stack.apis.inference import OpenAIChatCompletionRequestWithExtraBody, OpenAIUserMessageParam
12
- from llama_stack.apis.tools.rag_tool import (
10
+ from llama_stack.providers.utils.inference.prompt_adapter import (
11
+ interleaved_content_as_str,
12
+ )
13
+ from llama_stack_api import (
13
14
  DefaultRAGQueryGeneratorConfig,
15
+ InterleavedContent,
14
16
  LLMRAGQueryGeneratorConfig,
17
+ OpenAIChatCompletionRequestWithExtraBody,
18
+ OpenAIUserMessageParam,
15
19
  RAGQueryGenerator,
16
20
  RAGQueryGeneratorConfig,
17
21
  )
18
- from llama_stack.providers.utils.inference.prompt_adapter import (
19
- interleaved_content_as_str,
20
- )
21
22
 
22
23
 
23
24
  async def generate_rag_query(