synth-ai 0.2.14__py3-none-any.whl → 0.4.1__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.

Potentially problematic release.


This version of synth-ai might be problematic. Click here for more details.

Files changed (1091) hide show
  1. synth_ai/__init__.py +19 -40
  2. synth_ai/__main__.py +30 -3
  3. synth_ai/cli/__init__.py +105 -70
  4. synth_ai/cli/__main__.py +42 -0
  5. synth_ai/cli/_internal/__init__.py +5 -0
  6. synth_ai/cli/_internal/modal_wrapper.py +31 -0
  7. synth_ai/cli/_internal/storage.py +20 -0
  8. synth_ai/cli/_internal/typer_patch.py +47 -0
  9. synth_ai/cli/_internal/validate_task_app.py +29 -0
  10. synth_ai/cli/agents/__init__.py +17 -0
  11. synth_ai/cli/agents/claude.py +77 -0
  12. synth_ai/cli/agents/codex.py +265 -0
  13. synth_ai/cli/agents/opencode.py +253 -0
  14. synth_ai/cli/commands/__init__.py +18 -0
  15. synth_ai/cli/commands/artifacts/__init__.py +13 -0
  16. synth_ai/cli/commands/artifacts/client.py +119 -0
  17. synth_ai/cli/commands/artifacts/config.py +57 -0
  18. synth_ai/cli/commands/artifacts/core.py +24 -0
  19. synth_ai/cli/commands/artifacts/download.py +188 -0
  20. synth_ai/cli/commands/artifacts/export.py +186 -0
  21. synth_ai/cli/commands/artifacts/list.py +156 -0
  22. synth_ai/cli/commands/artifacts/parsing.py +250 -0
  23. synth_ai/cli/commands/artifacts/show.py +336 -0
  24. synth_ai/cli/commands/baseline/__init__.py +12 -0
  25. synth_ai/cli/commands/baseline/core.py +636 -0
  26. synth_ai/cli/commands/baseline/list.py +94 -0
  27. synth_ai/cli/commands/demo/__init__.py +3 -0
  28. synth_ai/cli/commands/demo/core.py +153 -0
  29. synth_ai/cli/commands/eval/__init__.py +19 -0
  30. synth_ai/cli/commands/eval/core.py +1113 -0
  31. synth_ai/cli/commands/eval/errors.py +81 -0
  32. synth_ai/cli/commands/eval/validation.py +133 -0
  33. synth_ai/cli/commands/filter/__init__.py +12 -0
  34. synth_ai/cli/commands/filter/core.py +424 -0
  35. synth_ai/cli/commands/filter/errors.py +55 -0
  36. synth_ai/cli/commands/filter/validation.py +77 -0
  37. synth_ai/cli/commands/help/__init__.py +185 -0
  38. synth_ai/cli/commands/help/core.py +72 -0
  39. synth_ai/cli/commands/scan/__init__.py +19 -0
  40. synth_ai/cli/commands/scan/cloudflare_scanner.py +403 -0
  41. synth_ai/cli/commands/scan/core.py +344 -0
  42. synth_ai/cli/commands/scan/health_checker.py +242 -0
  43. synth_ai/cli/commands/scan/local_scanner.py +278 -0
  44. synth_ai/cli/commands/scan/models.py +83 -0
  45. synth_ai/cli/commands/smoke/__init__.py +7 -0
  46. synth_ai/cli/commands/smoke/core.py +1438 -0
  47. synth_ai/cli/commands/status/__init__.py +66 -0
  48. synth_ai/cli/commands/status/client.py +192 -0
  49. synth_ai/cli/commands/status/config.py +92 -0
  50. synth_ai/cli/commands/status/errors.py +20 -0
  51. synth_ai/cli/commands/status/formatters.py +164 -0
  52. synth_ai/cli/commands/status/subcommands/__init__.py +9 -0
  53. synth_ai/cli/commands/status/subcommands/files.py +79 -0
  54. synth_ai/cli/commands/status/subcommands/jobs.py +334 -0
  55. synth_ai/cli/commands/status/subcommands/models.py +79 -0
  56. synth_ai/cli/commands/status/subcommands/pricing.py +23 -0
  57. synth_ai/cli/commands/status/subcommands/runs.py +81 -0
  58. synth_ai/cli/commands/status/subcommands/session.py +182 -0
  59. synth_ai/cli/commands/status/subcommands/summary.py +47 -0
  60. synth_ai/cli/commands/status/subcommands/usage.py +203 -0
  61. synth_ai/cli/commands/status/utils.py +114 -0
  62. synth_ai/cli/commands/train/__init__.py +53 -0
  63. synth_ai/cli/commands/train/core.py +22 -0
  64. synth_ai/cli/commands/train/errors.py +117 -0
  65. synth_ai/cli/commands/train/judge_schemas.py +201 -0
  66. synth_ai/cli/commands/train/judge_validation.py +305 -0
  67. synth_ai/cli/commands/train/prompt_learning_validation.py +633 -0
  68. synth_ai/cli/commands/train/validation.py +392 -0
  69. synth_ai/cli/demo_apps/__init__.py +10 -0
  70. synth_ai/cli/demo_apps/core/__init__.py +28 -0
  71. synth_ai/cli/demo_apps/core/cli.py +1735 -0
  72. synth_ai/cli/demo_apps/crafter/crafter_fft_4b.toml +55 -0
  73. synth_ai/cli/demo_apps/crafter/grpo_crafter_task_app.py +186 -0
  74. synth_ai/cli/demo_apps/crafter/rl_from_base_qwen4b.toml +74 -0
  75. synth_ai/cli/demo_apps/demo_registry.py +176 -0
  76. synth_ai/cli/demo_apps/demo_task_apps/core.py +440 -0
  77. synth_ai/cli/demo_apps/demo_task_apps/crafter/__init__.py +1 -0
  78. synth_ai/cli/demo_apps/demo_task_apps/crafter/grpo_crafter_task_app.py +185 -0
  79. synth_ai/cli/demo_apps/demo_task_apps/math/modal_task_app.py +742 -0
  80. synth_ai/cli/demo_apps/demo_task_apps/math/task_app_entry.py +39 -0
  81. synth_ai/cli/demo_apps/math/__init__.py +1 -0
  82. synth_ai/cli/demo_apps/math/_common.py +16 -0
  83. synth_ai/cli/demo_apps/math/app.py +38 -0
  84. synth_ai/cli/demo_apps/math/config.toml +76 -0
  85. synth_ai/cli/demo_apps/math/deploy_modal.py +54 -0
  86. synth_ai/cli/demo_apps/math/modal_task_app.py +702 -0
  87. synth_ai/cli/demo_apps/math/task_app_entry.py +53 -0
  88. synth_ai/cli/demo_apps/mipro/main.py +271 -0
  89. synth_ai/cli/demo_apps/mipro/task_app.py +933 -0
  90. synth_ai/cli/demo_apps/mipro/train_cfg.toml +92 -0
  91. synth_ai/cli/demos/__init__.py +12 -0
  92. synth_ai/cli/demos/demo.py +32 -0
  93. synth_ai/cli/demos/rl_demo.py +254 -0
  94. synth_ai/cli/deploy.py +216 -0
  95. synth_ai/cli/infra/__init__.py +14 -0
  96. synth_ai/cli/infra/balance.py +216 -0
  97. synth_ai/cli/infra/mcp.py +35 -0
  98. synth_ai/cli/infra/modal_app.py +36 -0
  99. synth_ai/cli/infra/setup.py +69 -0
  100. synth_ai/cli/infra/status.py +16 -0
  101. synth_ai/cli/infra/turso.py +77 -0
  102. synth_ai/cli/lib/__init__.py +10 -0
  103. synth_ai/cli/lib/agents.py +76 -0
  104. synth_ai/cli/lib/apps/modal_app.py +101 -0
  105. synth_ai/cli/lib/apps/task_app.py +643 -0
  106. synth_ai/cli/lib/bin.py +39 -0
  107. synth_ai/cli/lib/env.py +375 -0
  108. synth_ai/cli/lib/errors.py +85 -0
  109. synth_ai/cli/lib/modal.py +315 -0
  110. synth_ai/cli/lib/plotting.py +126 -0
  111. synth_ai/cli/lib/prompt_args.py +39 -0
  112. synth_ai/cli/lib/prompts.py +284 -0
  113. synth_ai/cli/lib/sqld.py +122 -0
  114. synth_ai/cli/lib/task_app_discovery.py +884 -0
  115. synth_ai/cli/lib/task_app_env.py +295 -0
  116. synth_ai/cli/lib/train_cfgs.py +300 -0
  117. synth_ai/cli/lib/tunnel_records.py +207 -0
  118. synth_ai/cli/local/__init__.py +14 -0
  119. synth_ai/cli/local/experiment_queue/__init__.py +72 -0
  120. synth_ai/cli/local/experiment_queue/api_schemas.py +221 -0
  121. synth_ai/cli/local/experiment_queue/celery_app.py +208 -0
  122. synth_ai/cli/local/experiment_queue/config.py +128 -0
  123. synth_ai/cli/local/experiment_queue/config_utils.py +272 -0
  124. synth_ai/cli/local/experiment_queue/database.py +175 -0
  125. synth_ai/cli/local/experiment_queue/dispatcher.py +119 -0
  126. synth_ai/cli/local/experiment_queue/models.py +231 -0
  127. synth_ai/cli/local/experiment_queue/progress_info.py +160 -0
  128. synth_ai/cli/local/experiment_queue/results.py +373 -0
  129. synth_ai/cli/local/experiment_queue/schemas.py +131 -0
  130. synth_ai/cli/local/experiment_queue/service.py +344 -0
  131. synth_ai/cli/local/experiment_queue/status.py +372 -0
  132. synth_ai/cli/local/experiment_queue/status_tracker.py +360 -0
  133. synth_ai/cli/local/experiment_queue/tasks.py +1984 -0
  134. synth_ai/cli/local/experiment_queue/trace_storage.py +65 -0
  135. synth_ai/cli/local/experiment_queue/validation.py +157 -0
  136. synth_ai/cli/local/session/__init__.py +92 -0
  137. synth_ai/cli/local/session/client.py +383 -0
  138. synth_ai/cli/local/session/constants.py +63 -0
  139. synth_ai/cli/local/session/exceptions.py +105 -0
  140. synth_ai/cli/local/session/manager.py +139 -0
  141. synth_ai/cli/local/session/models.py +89 -0
  142. synth_ai/cli/local/session/query.py +110 -0
  143. synth_ai/cli/root.py +30 -6
  144. synth_ai/cli/task_apps/__init__.py +26 -0
  145. synth_ai/cli/task_apps/commands.py +3153 -0
  146. synth_ai/cli/task_apps/deploy.py +7 -0
  147. synth_ai/cli/task_apps/list.py +26 -0
  148. synth_ai/cli/task_apps/main.py +36 -0
  149. synth_ai/cli/task_apps/modal_serve.py +11 -0
  150. synth_ai/cli/task_apps/serve.py +11 -0
  151. synth_ai/cli/training/__init__.py +8 -0
  152. synth_ai/cli/training/train.py +5 -0
  153. synth_ai/cli/training/train_cfg.py +34 -0
  154. synth_ai/cli/training/watch.py +506 -0
  155. synth_ai/cli/turso.py +34 -55
  156. synth_ai/cli/usage.py +159 -0
  157. synth_ai/cli/utils/__init__.py +8 -0
  158. synth_ai/cli/utils/experiments.py +235 -0
  159. synth_ai/cli/utils/queue.py +504 -0
  160. synth_ai/cli/utils/recent.py +133 -0
  161. synth_ai/cli/utils/traces.py +164 -0
  162. synth_ai/contracts/__init__.py +67 -0
  163. synth_ai/core/__init__.py +100 -0
  164. synth_ai/core/_utils/__init__.py +54 -0
  165. synth_ai/core/_utils/base_url.py +10 -0
  166. synth_ai/core/_utils/http.py +10 -0
  167. synth_ai/core/_utils/prompts.py +14 -0
  168. synth_ai/core/_utils/task_app_state.py +12 -0
  169. synth_ai/core/_utils/user_config.py +10 -0
  170. synth_ai/core/apps/common.py +116 -0
  171. synth_ai/core/auth.py +95 -0
  172. synth_ai/core/cfgs.py +240 -0
  173. synth_ai/core/config/__init__.py +16 -0
  174. synth_ai/core/config/base.py +168 -0
  175. synth_ai/core/config/resolver.py +89 -0
  176. synth_ai/core/env.py +220 -0
  177. synth_ai/core/errors.py +126 -0
  178. synth_ai/core/http.py +230 -0
  179. synth_ai/core/integrations/__init__.py +11 -0
  180. synth_ai/core/integrations/cloudflare.py +1710 -0
  181. synth_ai/core/integrations/mcp/__init__.py +6 -0
  182. synth_ai/core/integrations/mcp/__main__.py +8 -0
  183. synth_ai/core/integrations/mcp/claude.py +36 -0
  184. synth_ai/core/integrations/mcp/main.py +254 -0
  185. synth_ai/core/integrations/mcp/setup.py +100 -0
  186. synth_ai/core/integrations/modal.py +277 -0
  187. synth_ai/core/json.py +72 -0
  188. synth_ai/core/log_filter.py +99 -0
  189. synth_ai/core/logging.py +82 -0
  190. synth_ai/core/paths.py +107 -0
  191. synth_ai/core/pricing.py +109 -0
  192. synth_ai/core/process.py +233 -0
  193. synth_ai/core/ssl.py +25 -0
  194. synth_ai/core/storage/__init__.py +71 -0
  195. synth_ai/core/task_app_state.py +318 -0
  196. synth_ai/core/telemetry.py +282 -0
  197. synth_ai/core/tracing_v3/__init__.py +99 -0
  198. synth_ai/core/tracing_v3/abstractions.py +302 -0
  199. synth_ai/core/tracing_v3/config.py +229 -0
  200. synth_ai/core/tracing_v3/constants.py +21 -0
  201. synth_ai/core/tracing_v3/db_config.py +182 -0
  202. synth_ai/core/tracing_v3/decorators.py +401 -0
  203. synth_ai/core/tracing_v3/llm_call_record_helpers.py +437 -0
  204. synth_ai/core/tracing_v3/migration_helper.py +119 -0
  205. synth_ai/core/tracing_v3/session_tracer.py +542 -0
  206. synth_ai/core/tracing_v3/storage/base.py +211 -0
  207. synth_ai/core/tracing_v3/storage/config.py +109 -0
  208. synth_ai/core/tracing_v3/storage/factory.py +39 -0
  209. synth_ai/core/tracing_v3/trace_utils.py +326 -0
  210. synth_ai/core/tracing_v3/turso/daemon.py +278 -0
  211. synth_ai/core/tracing_v3/turso/models.py +470 -0
  212. synth_ai/core/tracing_v3/turso/native_manager.py +1385 -0
  213. synth_ai/core/tracing_v3/utils.py +108 -0
  214. synth_ai/core/urls.py +18 -0
  215. synth_ai/core/user_config.py +137 -0
  216. synth_ai/core/uvicorn.py +222 -0
  217. synth_ai/data/__init__.py +110 -0
  218. synth_ai/data/enums.py +141 -0
  219. synth_ai/data/rewards.py +152 -0
  220. synth_ai/data/specs.py +36 -0
  221. synth_ai/data/traces.py +35 -0
  222. synth_ai/products/__init__.py +6 -0
  223. synth_ai/products/graph_evolve/__init__.py +46 -0
  224. synth_ai/products/graph_evolve/client.py +226 -0
  225. synth_ai/products/graph_evolve/config.py +591 -0
  226. synth_ai/products/graph_evolve/converters/__init__.py +42 -0
  227. synth_ai/products/graph_evolve/converters/openai_sft.py +484 -0
  228. synth_ai/products/graph_evolve/examples/hotpotqa/config.toml +109 -0
  229. synth_ai/products/graph_evolve/run.py +222 -0
  230. synth_ai/sdk/__init__.py +119 -0
  231. synth_ai/sdk/api/__init__.py +1 -0
  232. synth_ai/sdk/api/models/supported.py +514 -0
  233. synth_ai/sdk/api/research_agent/__init__.py +86 -0
  234. synth_ai/sdk/api/research_agent/cli.py +428 -0
  235. synth_ai/sdk/api/research_agent/config.py +357 -0
  236. synth_ai/sdk/api/research_agent/job.py +717 -0
  237. synth_ai/sdk/api/train/__init__.py +85 -0
  238. synth_ai/sdk/api/train/builders.py +895 -0
  239. synth_ai/sdk/api/train/cli.py +2188 -0
  240. synth_ai/sdk/api/train/config_finder.py +267 -0
  241. synth_ai/sdk/api/train/configs/__init__.py +65 -0
  242. synth_ai/sdk/api/train/configs/prompt_learning.py +1706 -0
  243. synth_ai/sdk/api/train/configs/rl.py +188 -0
  244. synth_ai/sdk/api/train/configs/sft.py +99 -0
  245. synth_ai/sdk/api/train/configs/shared.py +81 -0
  246. synth_ai/sdk/api/train/context_learning.py +312 -0
  247. synth_ai/sdk/api/train/env_resolver.py +418 -0
  248. synth_ai/sdk/api/train/graph_validators.py +216 -0
  249. synth_ai/sdk/api/train/graphgen.py +984 -0
  250. synth_ai/sdk/api/train/graphgen_models.py +823 -0
  251. synth_ai/sdk/api/train/graphgen_validators.py +109 -0
  252. synth_ai/sdk/api/train/pollers.py +124 -0
  253. synth_ai/sdk/api/train/progress/__init__.py +97 -0
  254. synth_ai/sdk/api/train/progress/dataclasses.py +569 -0
  255. synth_ai/sdk/api/train/progress/events.py +326 -0
  256. synth_ai/sdk/api/train/progress/results.py +428 -0
  257. synth_ai/sdk/api/train/progress/tracker.py +641 -0
  258. synth_ai/sdk/api/train/prompt_learning.py +470 -0
  259. synth_ai/sdk/api/train/rl.py +442 -0
  260. synth_ai/sdk/api/train/sft.py +396 -0
  261. synth_ai/sdk/api/train/summary.py +522 -0
  262. synth_ai/sdk/api/train/supported_algos.py +147 -0
  263. synth_ai/sdk/api/train/task_app.py +331 -0
  264. synth_ai/sdk/api/train/utils.py +279 -0
  265. synth_ai/sdk/api/train/validators.py +2424 -0
  266. synth_ai/sdk/baseline/__init__.py +25 -0
  267. synth_ai/sdk/baseline/config.py +209 -0
  268. synth_ai/sdk/baseline/discovery.py +216 -0
  269. synth_ai/sdk/baseline/execution.py +154 -0
  270. synth_ai/sdk/graphs/__init__.py +15 -0
  271. synth_ai/sdk/graphs/completions.py +570 -0
  272. synth_ai/sdk/inference/__init__.py +6 -0
  273. synth_ai/sdk/inference/client.py +128 -0
  274. synth_ai/sdk/jobs/__init__.py +16 -0
  275. synth_ai/sdk/jobs/client.py +371 -0
  276. synth_ai/sdk/judging/__init__.py +15 -0
  277. synth_ai/sdk/judging/base.py +24 -0
  278. synth_ai/sdk/judging/client.py +191 -0
  279. synth_ai/sdk/judging/schemas.py +222 -0
  280. synth_ai/sdk/learning/__init__.py +69 -0
  281. synth_ai/sdk/learning/client.py +240 -0
  282. synth_ai/sdk/learning/ft_client.py +7 -0
  283. synth_ai/sdk/learning/health.py +49 -0
  284. synth_ai/sdk/learning/jobs.py +202 -0
  285. synth_ai/sdk/learning/prompt_extraction.py +334 -0
  286. synth_ai/sdk/learning/prompt_learning_client.py +455 -0
  287. synth_ai/sdk/learning/prompt_learning_types.py +185 -0
  288. synth_ai/sdk/learning/rl/client.py +268 -0
  289. synth_ai/sdk/learning/rl/contracts.py +27 -0
  290. synth_ai/sdk/learning/rl/env_keys.py +166 -0
  291. synth_ai/sdk/learning/rl/secrets.py +13 -0
  292. synth_ai/sdk/learning/sft/client.py +95 -0
  293. synth_ai/sdk/learning/sft/config.py +270 -0
  294. synth_ai/sdk/learning/sft/data.py +698 -0
  295. synth_ai/sdk/learning/validators.py +52 -0
  296. synth_ai/sdk/research_agent/__init__.py +34 -0
  297. synth_ai/sdk/research_agent/container_builder.py +328 -0
  298. synth_ai/sdk/research_agent/container_spec.py +198 -0
  299. synth_ai/sdk/research_agent/defaults.py +34 -0
  300. synth_ai/sdk/research_agent/results_collector.py +69 -0
  301. synth_ai/sdk/specs/__init__.py +46 -0
  302. synth_ai/sdk/specs/dataclasses.py +149 -0
  303. synth_ai/sdk/specs/loader.py +144 -0
  304. synth_ai/sdk/specs/serializer.py +199 -0
  305. synth_ai/sdk/specs/validation.py +250 -0
  306. synth_ai/sdk/streaming/__init__.py +35 -0
  307. synth_ai/sdk/streaming/config.py +94 -0
  308. synth_ai/sdk/streaming/handlers.py +1997 -0
  309. synth_ai/sdk/streaming/streamer.py +704 -0
  310. synth_ai/sdk/streaming/types.py +112 -0
  311. synth_ai/sdk/task/__init__.py +151 -0
  312. synth_ai/sdk/task/apps/__init__.py +133 -0
  313. synth_ai/sdk/task/config.py +261 -0
  314. synth_ai/sdk/task/contracts.py +298 -0
  315. synth_ai/sdk/task/datasets.py +108 -0
  316. synth_ai/sdk/task/in_process.py +1190 -0
  317. synth_ai/sdk/task/in_process_runner.py +309 -0
  318. synth_ai/sdk/task/inference_api.py +299 -0
  319. synth_ai/sdk/task/proxy.py +287 -0
  320. synth_ai/sdk/task/rubrics/__init__.py +55 -0
  321. synth_ai/sdk/task/rubrics/loaders.py +156 -0
  322. synth_ai/sdk/task/rubrics.py +219 -0
  323. synth_ai/sdk/task/server.py +580 -0
  324. synth_ai/sdk/task/trace_correlation_helpers.py +506 -0
  325. synth_ai/sdk/task/tracing_utils.py +95 -0
  326. synth_ai/sdk/task/validators.py +456 -0
  327. synth_ai/sdk/tracing/__init__.py +39 -0
  328. synth_ai/sdk/training/__init__.py +102 -0
  329. synth_ai/sdk/usage/__init__.py +37 -0
  330. synth_ai/sdk/usage/client.py +171 -0
  331. synth_ai/sdk/usage/models.py +261 -0
  332. synth_ai/utils/__init__.py +213 -0
  333. synth_ai-0.4.1.dist-info/METADATA +195 -0
  334. synth_ai-0.4.1.dist-info/RECORD +379 -0
  335. synth_ai-0.4.1.dist-info/top_level.txt +1 -0
  336. examples/__init__.py +0 -16
  337. examples/analyze_semantic_words.sh +0 -17
  338. examples/crafter_debug_render.py +0 -186
  339. examples/dev/qwen3_32b_qlora_4xh100.toml +0 -40
  340. examples/multi_step/configs/README_verilog_rl.md +0 -77
  341. examples/multi_step/configs/VERILOG_REWARDS.md +0 -90
  342. examples/multi_step/configs/VERILOG_RL_CHECKLIST.md +0 -183
  343. examples/multi_step/configs/crafter_eval_synth_qwen4b.toml +0 -35
  344. examples/multi_step/configs/crafter_eval_text_only_groq_qwen32b.toml +0 -36
  345. examples/multi_step/configs/crafter_rl_outcome.toml +0 -74
  346. examples/multi_step/configs/crafter_rl_stepwise_hosted_judge.toml +0 -187
  347. examples/multi_step/configs/crafter_rl_stepwise_shaped.toml +0 -83
  348. examples/multi_step/configs/crafter_rl_stepwise_simple.toml +0 -78
  349. examples/multi_step/configs/crafter_synth_backend.md +0 -40
  350. examples/multi_step/configs/verilog_eval_groq_qwen32b.toml +0 -31
  351. examples/multi_step/configs/verilog_eval_synth_qwen8b.toml +0 -33
  352. examples/multi_step/configs/verilog_rl_lora.toml +0 -190
  353. examples/multi_step/crafter_rl_lora.md +0 -70
  354. examples/multi_step/judges/crafter_backend_judge.py +0 -220
  355. examples/multi_step/judges/verilog_backend_judge.py +0 -234
  356. examples/multi_step/readme.md +0 -48
  357. examples/multi_step/sse_metrics_streaming_notes.md +0 -357
  358. examples/multi_step/task_app_config_notes.md +0 -494
  359. examples/multi_step/verilog_rl_lora.md +0 -218
  360. examples/qwen_coder/README.md +0 -102
  361. examples/qwen_coder/_shared.py +0 -113
  362. examples/qwen_coder/configs/coder_lora_30b.toml +0 -61
  363. examples/qwen_coder/configs/coder_lora_4b.toml +0 -57
  364. examples/qwen_coder/configs/coder_lora_small.toml +0 -58
  365. examples/qwen_coder/generate_dataset.py +0 -98
  366. examples/qwen_coder/infer_ft_smoke.py +0 -65
  367. examples/qwen_coder/infer_prod_proxy.py +0 -73
  368. examples/qwen_coder/infer_via_synth.py +0 -87
  369. examples/qwen_coder/scripts/infer_coder.sh +0 -19
  370. examples/qwen_coder/scripts/train_coder_30b.sh +0 -22
  371. examples/qwen_coder/sft_full_17b.py +0 -103
  372. examples/qwen_coder/sft_lora_30b.py +0 -110
  373. examples/qwen_coder/subset_jsonl.py +0 -39
  374. examples/qwen_coder/todos.md +0 -38
  375. examples/qwen_coder/validate_jsonl.py +0 -60
  376. examples/rl/README.md +0 -169
  377. examples/rl/download_dataset.py +0 -80
  378. examples/run_crafter_demo.sh +0 -10
  379. examples/sft/README.md +0 -139
  380. examples/sft/configs/crafter_fft_qwen0p6b.toml +0 -44
  381. examples/sft/configs/crafter_lora_qwen0p6b.toml +0 -45
  382. examples/sft/evaluate.py +0 -119
  383. examples/sft/export_dataset.py +0 -117
  384. examples/sft/generate_traces.py +0 -164
  385. examples/swe/__init__.py +0 -12
  386. examples/swe/task_app/README.md +0 -105
  387. examples/swe/task_app/__init__.py +0 -2
  388. examples/swe/task_app/grpo_swe_mini.py +0 -601
  389. examples/swe/task_app/grpo_swe_mini_task_app.py +0 -136
  390. examples/swe/task_app/hosted/README.md +0 -173
  391. examples/swe/task_app/hosted/__init__.py +0 -5
  392. examples/swe/task_app/hosted/branching.py +0 -143
  393. examples/swe/task_app/hosted/environment_routes.py +0 -1289
  394. examples/swe/task_app/hosted/envs/__init__.py +0 -1
  395. examples/swe/task_app/hosted/envs/crafter/__init__.py +0 -6
  396. examples/swe/task_app/hosted/envs/crafter/app.py +0 -1
  397. examples/swe/task_app/hosted/envs/crafter/environment.py +0 -522
  398. examples/swe/task_app/hosted/envs/crafter/policy.py +0 -478
  399. examples/swe/task_app/hosted/envs/crafter/react_agent.py +0 -108
  400. examples/swe/task_app/hosted/envs/crafter/shared.py +0 -305
  401. examples/swe/task_app/hosted/envs/crafter/tools.py +0 -47
  402. examples/swe/task_app/hosted/envs/mini_swe/__init__.py +0 -8
  403. examples/swe/task_app/hosted/envs/mini_swe/environment.py +0 -1164
  404. examples/swe/task_app/hosted/envs/mini_swe/policy.py +0 -355
  405. examples/swe/task_app/hosted/envs/mini_swe/shared.py +0 -83
  406. examples/swe/task_app/hosted/envs/mini_swe/tools.py +0 -96
  407. examples/swe/task_app/hosted/hosted_app.py +0 -204
  408. examples/swe/task_app/hosted/inference/__init__.py +0 -5
  409. examples/swe/task_app/hosted/inference/openai_client.py +0 -618
  410. examples/swe/task_app/hosted/main.py +0 -100
  411. examples/swe/task_app/hosted/policy_routes.py +0 -1079
  412. examples/swe/task_app/hosted/registry.py +0 -195
  413. examples/swe/task_app/hosted/rollout.py +0 -1911
  414. examples/swe/task_app/hosted/storage/__init__.py +0 -5
  415. examples/swe/task_app/hosted/storage/volume.py +0 -211
  416. examples/swe/task_app/hosted/test_agents.py +0 -161
  417. examples/swe/task_app/hosted/test_service.py +0 -136
  418. examples/swe/task_app/hosted/utils.py +0 -62
  419. examples/task_apps/IMAGE_ONLY_EVAL_QUICKSTART.md +0 -258
  420. examples/task_apps/TESTING.md +0 -275
  421. examples/task_apps/crafter/CREATE_SFT_DATASET.md +0 -273
  422. examples/task_apps/crafter/EVAL_IMAGE_ONLY_RESULTS.md +0 -152
  423. examples/task_apps/crafter/FILTER_COMMAND_STATUS.md +0 -174
  424. examples/task_apps/crafter/FILTER_COMMAND_SUCCESS.md +0 -268
  425. examples/task_apps/crafter/QUERY_EXAMPLES.md +0 -203
  426. examples/task_apps/crafter/README_IMAGE_ONLY_EVAL.md +0 -316
  427. examples/task_apps/crafter/__init__.py +0 -0
  428. examples/task_apps/crafter/eval_image_only_gpt4o.toml +0 -28
  429. examples/task_apps/crafter/eval_text_only_groq_llama.toml +0 -36
  430. examples/task_apps/crafter/filter_sft_dataset.toml +0 -16
  431. examples/task_apps/crafter/task_app/README.md +0 -42
  432. examples/task_apps/crafter/task_app/__init__.py +0 -5
  433. examples/task_apps/crafter/task_app/grpo_crafter.py +0 -973
  434. examples/task_apps/crafter/task_app/grpo_crafter_task_app.py +0 -146
  435. examples/task_apps/crafter/task_app/synth_envs_hosted/README.md +0 -173
  436. examples/task_apps/crafter/task_app/synth_envs_hosted/__init__.py +0 -5
  437. examples/task_apps/crafter/task_app/synth_envs_hosted/branching.py +0 -143
  438. examples/task_apps/crafter/task_app/synth_envs_hosted/environment_routes.py +0 -1226
  439. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/__init__.py +0 -1
  440. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/__init__.py +0 -6
  441. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/app.py +0 -1
  442. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/environment.py +0 -532
  443. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/policy.py +0 -547
  444. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/react_agent.py +0 -123
  445. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/shared.py +0 -305
  446. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/tools.py +0 -47
  447. examples/task_apps/crafter/task_app/synth_envs_hosted/hosted_app.py +0 -204
  448. examples/task_apps/crafter/task_app/synth_envs_hosted/inference/__init__.py +0 -5
  449. examples/task_apps/crafter/task_app/synth_envs_hosted/inference/openai_client.py +0 -704
  450. examples/task_apps/crafter/task_app/synth_envs_hosted/main.py +0 -100
  451. examples/task_apps/crafter/task_app/synth_envs_hosted/policy_routes.py +0 -1152
  452. examples/task_apps/crafter/task_app/synth_envs_hosted/registry.py +0 -195
  453. examples/task_apps/crafter/task_app/synth_envs_hosted/rollout.py +0 -2160
  454. examples/task_apps/crafter/task_app/synth_envs_hosted/storage/__init__.py +0 -5
  455. examples/task_apps/crafter/task_app/synth_envs_hosted/storage/volume.py +0 -211
  456. examples/task_apps/crafter/task_app/synth_envs_hosted/test_agents.py +0 -161
  457. examples/task_apps/crafter/task_app/synth_envs_hosted/test_service.py +0 -136
  458. examples/task_apps/crafter/task_app/synth_envs_hosted/utils.py +0 -218
  459. examples/task_apps/dev/pokemon_emerald/__init__.py +0 -2
  460. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/README.md +0 -811
  461. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/__init__.py +0 -120
  462. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/action.py +0 -160
  463. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/memory.py +0 -155
  464. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/perception.py +0 -69
  465. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/planning.py +0 -96
  466. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/simple.py +0 -1502
  467. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/system_prompt.py +0 -4
  468. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/grab_map.py +0 -68
  469. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/manual.py +0 -216
  470. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/__init__.py +0 -35
  471. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/emerald_utils.py +0 -631
  472. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/emulator.py +0 -1544
  473. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/enums.py +0 -1428
  474. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/memory_reader.py +0 -4848
  475. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/types.py +0 -41
  476. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/utils.py +0 -298
  477. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pyproject.toml +0 -95
  478. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/run.py +0 -204
  479. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/server/__init__.py +0 -0
  480. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/server/app.py +0 -2152
  481. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/server/client.py +0 -429
  482. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/server/frame_server.py +0 -155
  483. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/README.md +0 -78
  484. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/__init__.py +0 -0
  485. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/run_tests.py +0 -122
  486. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_agent_direct.py +0 -76
  487. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_agent_prompts.py +0 -413
  488. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_battle_state_formatting.py +0 -204
  489. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_dialogue_detection.py +0 -133
  490. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_dialogue_detection_comprehensive.py +0 -229
  491. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_direct_agent_emulator.py +0 -300
  492. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_fps_adjustment_pytest.py +0 -205
  493. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_house_to_outside_direct.py +0 -200
  494. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_house_to_outside_transition.py +0 -284
  495. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_map_ground_truth_comparison.py +0 -468
  496. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_memory_map.py +0 -575
  497. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_server_map_validation.py +0 -311
  498. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_torchic_state.py +0 -259
  499. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/__init__.py +0 -0
  500. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/anticheat.py +0 -372
  501. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/checkpoint.py +0 -296
  502. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/error_handler.py +0 -275
  503. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/get_local_ip.py +0 -22
  504. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/helpers.py +0 -44
  505. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/llm_logger.py +0 -514
  506. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_formatter.py +0 -415
  507. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_stitcher.py +0 -1763
  508. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_stitcher_singleton.py +0 -33
  509. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_trimmer.py +0 -106
  510. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_visualizer.py +0 -334
  511. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/ocr_dialogue.py +0 -1020
  512. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/recording.py +0 -188
  513. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/state_formatter.py +0 -1481
  514. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/vlm.py +0 -862
  515. examples/task_apps/dev/pokemon_emerald/modal_app.py +0 -114
  516. examples/task_apps/dev/pokemon_emerald/task_app/README.md +0 -81
  517. examples/task_apps/dev/pokemon_emerald/task_app/__init__.py +0 -6
  518. examples/task_apps/dev/pokemon_emerald/task_app/pokemon_emerald.py +0 -685
  519. examples/task_apps/enron/__init__.py +0 -1
  520. examples/task_apps/enron/eval_groq_qwen32.toml +0 -16
  521. examples/task_apps/enron/filter_sft.toml +0 -5
  522. examples/task_apps/enron/task_app/README.md +0 -14
  523. examples/task_apps/enron/task_app/__init__.py +0 -1
  524. examples/task_apps/enron/task_app/grpo_enron.py +0 -906
  525. examples/task_apps/enron/task_app/grpo_enron_task_app.py +0 -146
  526. examples/task_apps/enron/tests/__init__.py +0 -4
  527. examples/task_apps/enron/tests/conftest.py +0 -115
  528. examples/task_apps/enron/tests/integration/__init__.py +0 -4
  529. examples/task_apps/enron/tests/integration/test_enron_eval.py +0 -179
  530. examples/task_apps/enron/tests/integration/test_enron_rollout.py +0 -135
  531. examples/task_apps/enron/tests/unit/__init__.py +0 -4
  532. examples/task_apps/enron/tests/unit/test_enron_environment.py +0 -126
  533. examples/task_apps/math/README.md +0 -22
  534. examples/task_apps/math/__init__.py +0 -0
  535. examples/task_apps/math/math_single_step.py +0 -1000
  536. examples/task_apps/math/math_task_app.py +0 -115
  537. examples/task_apps/pokemon_battle/__init__.py +0 -2
  538. examples/task_apps/pokemon_battle/modal_app.py +0 -104
  539. examples/task_apps/pokemon_battle/task_app/README.md +0 -68
  540. examples/task_apps/pokemon_battle/task_app/__init__.py +0 -6
  541. examples/task_apps/pokemon_battle/task_app/pokemon_showdown.py +0 -932
  542. examples/task_apps/pokemon_red/EVAL_IMAGE_ONLY_COMPLETE.md +0 -283
  543. examples/task_apps/pokemon_red/EVAL_IMAGE_ONLY_STATUS.md +0 -155
  544. examples/task_apps/pokemon_red/README.md +0 -357
  545. examples/task_apps/pokemon_red/README_IMAGE_ONLY_EVAL.md +0 -415
  546. examples/task_apps/pokemon_red/__init__.py +0 -3
  547. examples/task_apps/pokemon_red/eval_image_only_gpt4o.toml +0 -29
  548. examples/task_apps/pokemon_red/eval_pokemon_red_policy.py +0 -225
  549. examples/task_apps/pokemon_red/pallet_town_rl_config.toml +0 -75
  550. examples/task_apps/pokemon_red/task_app.py +0 -799
  551. examples/task_apps/pokemon_red/test_pallet_town_rewards.py +0 -193
  552. examples/task_apps/sokoban/README.md +0 -307
  553. examples/task_apps/sokoban/__init__.py +0 -3
  554. examples/task_apps/sokoban/eval_groq_qwen32.toml +0 -16
  555. examples/task_apps/sokoban/eval_openai_gpt5.toml +0 -16
  556. examples/task_apps/sokoban/filter_sft.toml +0 -5
  557. examples/task_apps/sokoban/task_app.py +0 -1058
  558. examples/task_apps/sokoban/tests/__init__.py +0 -4
  559. examples/task_apps/sokoban/tests/conftest.py +0 -113
  560. examples/task_apps/sokoban/tests/integration/__init__.py +0 -4
  561. examples/task_apps/sokoban/tests/integration/test_sokoban_eval.py +0 -57
  562. examples/task_apps/sokoban/tests/integration/test_sokoban_rollout.py +0 -198
  563. examples/task_apps/sokoban/tests/unit/__init__.py +0 -4
  564. examples/task_apps/sokoban/tests/unit/test_sokoban_environment.py +0 -114
  565. examples/task_apps/verilog/__init__.py +0 -1
  566. examples/task_apps/verilog/eval_groq_qwen32b.toml +0 -24
  567. examples/task_apps/verilog/filter_sft.toml +0 -5
  568. examples/task_apps/verilog/task_app/README.md +0 -12
  569. examples/task_apps/verilog/task_app/__init__.py +0 -1
  570. examples/task_apps/verilog/task_app/grpo_verilog.py +0 -1166
  571. examples/task_apps/verilog/task_app/grpo_verilog_task_app.py +0 -145
  572. examples/task_apps/verilog/tests/__init__.py +0 -4
  573. examples/task_apps/verilog/tests/conftest.py +0 -115
  574. examples/task_apps/verilog/tests/integration/__init__.py +0 -4
  575. examples/task_apps/verilog/tests/integration/test_verilog_eval.py +0 -181
  576. examples/task_apps/verilog/tests/integration/test_verilog_rollout.py +0 -55
  577. examples/task_apps/verilog/tests/unit/__init__.py +0 -4
  578. examples/task_apps/verilog/tests/unit/test_verilog_scoring.py +0 -118
  579. examples/vlm/PROPOSAL.md +0 -53
  580. examples/vlm/README.md +0 -68
  581. examples/vlm/configs/crafter_vlm_gpt4o.toml +0 -44
  582. examples/vlm/crafter_image_only_agent.py +0 -207
  583. examples/vlm/crafter_openai_vlm_agent.py +0 -277
  584. examples/vlm/filter_image_rows.py +0 -63
  585. examples/vlm/run_crafter_vlm_benchmark.py +0 -316
  586. examples/warming_up_to_rl/analyze_trace_db.py +0 -422
  587. examples/warming_up_to_rl/configs/crafter_fft.toml +0 -48
  588. examples/warming_up_to_rl/configs/crafter_fft_4b.toml +0 -54
  589. examples/warming_up_to_rl/configs/eval_fft_qwen4b.toml +0 -20
  590. examples/warming_up_to_rl/configs/eval_groq_qwen32b.toml +0 -13
  591. examples/warming_up_to_rl/configs/eval_modal_qwen4b.toml +0 -23
  592. examples/warming_up_to_rl/configs/eval_stepwise_complex.toml +0 -35
  593. examples/warming_up_to_rl/configs/eval_stepwise_consistent.toml +0 -26
  594. examples/warming_up_to_rl/configs/eval_stepwise_per_achievement.toml +0 -36
  595. examples/warming_up_to_rl/configs/eval_stepwise_simple.toml +0 -32
  596. examples/warming_up_to_rl/configs/rl_from_base_qwen4b.toml +0 -83
  597. examples/warming_up_to_rl/configs/rl_from_ft.toml +0 -56
  598. examples/warming_up_to_rl/export_trace_sft.py +0 -723
  599. examples/warming_up_to_rl/groq_test.py +0 -97
  600. examples/warming_up_to_rl/manage_secrets.py +0 -131
  601. examples/warming_up_to_rl/old/event_rewards.md +0 -234
  602. examples/warming_up_to_rl/old/notes.md +0 -73
  603. examples/warming_up_to_rl/readme.md +0 -179
  604. examples/warming_up_to_rl/run_eval.py +0 -736
  605. examples/warming_up_to_rl/run_fft_and_save.py +0 -380
  606. examples/warming_up_to_rl/run_local_rollout.py +0 -239
  607. examples/warming_up_to_rl/run_local_rollout_modal.py +0 -248
  608. examples/warming_up_to_rl/run_local_rollout_parallel.py +0 -405
  609. examples/warming_up_to_rl/run_local_rollout_traced.py +0 -477
  610. examples/warming_up_to_rl/run_rl_and_save.py +0 -124
  611. examples/warming_up_to_rl/run_rollout_remote.py +0 -156
  612. examples/workflows/__init__.py +0 -0
  613. examples/workflows/math_rl/__init__.py +0 -0
  614. examples/workflows/math_rl/configs/eval_base_qwen.toml +0 -15
  615. examples/workflows/math_rl/configs/eval_rl_qwen.toml +0 -11
  616. examples/workflows/math_rl/configs/rl_from_base_qwen.toml +0 -35
  617. examples/workflows/math_rl/configs/rl_from_base_qwen17.toml +0 -74
  618. examples/workflows/math_rl/configs/rl_from_ft_qwen.toml +0 -35
  619. examples/workflows/math_rl/download_dataset.py +0 -80
  620. examples/workflows/math_rl/run_eval.py +0 -436
  621. examples/workflows/math_rl/run_rl_and_save.py +0 -111
  622. synth_ai/api/models/supported.py +0 -377
  623. synth_ai/api/train/__init__.py +0 -5
  624. synth_ai/api/train/builders.py +0 -351
  625. synth_ai/api/train/cli.py +0 -635
  626. synth_ai/api/train/config_finder.py +0 -228
  627. synth_ai/api/train/configs/__init__.py +0 -44
  628. synth_ai/api/train/configs/rl.py +0 -134
  629. synth_ai/api/train/configs/sft.py +0 -95
  630. synth_ai/api/train/configs/shared.py +0 -24
  631. synth_ai/api/train/env_resolver.py +0 -349
  632. synth_ai/api/train/pollers.py +0 -75
  633. synth_ai/api/train/supported_algos.py +0 -147
  634. synth_ai/api/train/task_app.py +0 -195
  635. synth_ai/api/train/utils.py +0 -225
  636. synth_ai/cli/_modal_wrapper.py +0 -29
  637. synth_ai/cli/_storage.py +0 -20
  638. synth_ai/cli/_typer_patch.py +0 -49
  639. synth_ai/cli/_validate_task_app.py +0 -11
  640. synth_ai/cli/balance.py +0 -216
  641. synth_ai/cli/calc.py +0 -84
  642. synth_ai/cli/demo.py +0 -165
  643. synth_ai/cli/legacy_root_backup.py +0 -468
  644. synth_ai/cli/man.py +0 -106
  645. synth_ai/cli/recent.py +0 -132
  646. synth_ai/cli/rl_demo.py +0 -254
  647. synth_ai/cli/status.py +0 -134
  648. synth_ai/cli/task_apps.py +0 -4523
  649. synth_ai/cli/traces.py +0 -164
  650. synth_ai/cli/tui.py +0 -57
  651. synth_ai/cli/watch.py +0 -506
  652. synth_ai/compound/cais.py +0 -0
  653. synth_ai/config/base_url.py +0 -107
  654. synth_ai/core/experiment.py +0 -13
  655. synth_ai/core/system.py +0 -15
  656. synth_ai/demo_registry.py +0 -295
  657. synth_ai/demos/core/__init__.py +0 -1
  658. synth_ai/demos/core/cli.py +0 -1718
  659. synth_ai/demos/demo_task_apps/core.py +0 -440
  660. synth_ai/demos/demo_task_apps/crafter/grpo_crafter_task_app.py +0 -184
  661. synth_ai/demos/demo_task_apps/math/deploy_task_app.sh +0 -22
  662. synth_ai/demos/demo_task_apps/math/modal_task_app.py +0 -739
  663. synth_ai/demos/demo_task_apps/math/task_app_entry.py +0 -37
  664. synth_ai/environments/__init__.py +0 -31
  665. synth_ai/environments/environment/__init__.py +0 -1
  666. synth_ai/environments/environment/artifacts/__init__.py +0 -1
  667. synth_ai/environments/environment/artifacts/base.py +0 -52
  668. synth_ai/environments/environment/core.py +0 -67
  669. synth_ai/environments/environment/db/__init__.py +0 -1
  670. synth_ai/environments/environment/db/sqlite.py +0 -45
  671. synth_ai/environments/environment/registry.py +0 -233
  672. synth_ai/environments/environment/resources/sqlite.py +0 -45
  673. synth_ai/environments/environment/results.py +0 -1
  674. synth_ai/environments/environment/rewards/__init__.py +0 -1
  675. synth_ai/environments/environment/rewards/core.py +0 -29
  676. synth_ai/environments/environment/shared_engine.py +0 -26
  677. synth_ai/environments/environment/tools/__init__.py +0 -200
  678. synth_ai/environments/examples/__init__.py +0 -1
  679. synth_ai/environments/examples/bandit/__init__.py +0 -33
  680. synth_ai/environments/examples/bandit/engine.py +0 -302
  681. synth_ai/environments/examples/bandit/environment.py +0 -194
  682. synth_ai/environments/examples/bandit/taskset.py +0 -200
  683. synth_ai/environments/examples/crafter_classic/__init__.py +0 -8
  684. synth_ai/environments/examples/crafter_classic/agent_demos/analyze_semantic_words_markdown.py +0 -250
  685. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_comprehensive_evaluation.py +0 -59
  686. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_evaluation_browser.py +0 -152
  687. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_evaluation_config.toml +0 -24
  688. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_evaluation_framework.py +0 -1194
  689. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/crafter_synth_config.toml +0 -56
  690. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/filter_config_modal.toml +0 -32
  691. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/filter_traces_sft_turso.py +0 -738
  692. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/kick_off_ft_modal.py +0 -384
  693. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_action_results.py +0 -53
  694. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_agent_actions.py +0 -178
  695. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_latest_run.py +0 -222
  696. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_lm_traces.py +0 -183
  697. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_no_rewards.py +0 -210
  698. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_trace_issue.py +0 -206
  699. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/check_db_schema.py +0 -49
  700. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/check_latest_results.py +0 -64
  701. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/debug_agent_responses.py +0 -88
  702. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/quick_trace_check.py +0 -77
  703. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/compare_experiments.py +0 -324
  704. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/filter_traces_sft_turso.py +0 -580
  705. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/kick_off_ft_oai.py +0 -362
  706. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/multi_model_config.toml +0 -49
  707. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/analyze_enhanced_hooks.py +0 -332
  708. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/analyze_hook_events.py +0 -97
  709. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/analyze_hook_results.py +0 -217
  710. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/check_hook_storage.py +0 -87
  711. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/check_seeds.py +0 -88
  712. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/compare_seed_performance.py +0 -195
  713. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/custom_eval_pipelines.py +0 -400
  714. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/plot_hook_frequency.py +0 -195
  715. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/seed_analysis_summary.py +0 -56
  716. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/run_rollouts_for_models_and_compare_v3.py +0 -858
  717. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_quick_evaluation.py +0 -52
  718. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_react_agent.py +0 -874
  719. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_trace_evaluation.py +0 -1412
  720. synth_ai/environments/examples/crafter_classic/agent_demos/example_v3_usage.py +0 -216
  721. synth_ai/environments/examples/crafter_classic/agent_demos/old/compare_traces.py +0 -296
  722. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_comprehensive_evaluation.py +0 -58
  723. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_env_serialization.py +0 -464
  724. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_evaluation_browser.py +0 -152
  725. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_quick_evaluation.py +0 -51
  726. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_trace_evaluation.py +0 -1412
  727. synth_ai/environments/examples/crafter_classic/agent_demos/old/debug_player_loss.py +0 -112
  728. synth_ai/environments/examples/crafter_classic/agent_demos/old/diagnose_service.py +0 -203
  729. synth_ai/environments/examples/crafter_classic/agent_demos/old/diagnose_slowness.py +0 -305
  730. synth_ai/environments/examples/crafter_classic/agent_demos/old/eval_by_difficulty.py +0 -126
  731. synth_ai/environments/examples/crafter_classic/agent_demos/old/eval_example.py +0 -94
  732. synth_ai/environments/examples/crafter_classic/agent_demos/old/explore_saved_states.py +0 -142
  733. synth_ai/environments/examples/crafter_classic/agent_demos/old/filter_traces_sft.py +0 -26
  734. synth_ai/environments/examples/crafter_classic/agent_demos/old/filter_traces_sft_OLD.py +0 -984
  735. synth_ai/environments/examples/crafter_classic/agent_demos/old/generate_ft_data_gemini.py +0 -724
  736. synth_ai/environments/examples/crafter_classic/agent_demos/old/generate_ft_data_modal.py +0 -386
  737. synth_ai/environments/examples/crafter_classic/agent_demos/old/generate_ft_metadata.py +0 -205
  738. synth_ai/environments/examples/crafter_classic/agent_demos/old/kick_off_ft_gemini.py +0 -150
  739. synth_ai/environments/examples/crafter_classic/agent_demos/old/kick_off_ft_modal.py +0 -283
  740. synth_ai/environments/examples/crafter_classic/agent_demos/old/prepare_vertex_ft.py +0 -280
  741. synth_ai/environments/examples/crafter_classic/agent_demos/old/profile_env_slowness.py +0 -456
  742. synth_ai/environments/examples/crafter_classic/agent_demos/old/replicate_issue.py +0 -166
  743. synth_ai/environments/examples/crafter_classic/agent_demos/old/run_and_eval.py +0 -102
  744. synth_ai/environments/examples/crafter_classic/agent_demos/old/run_comparison.py +0 -128
  745. synth_ai/environments/examples/crafter_classic/agent_demos/old/run_qwen_rollouts.py +0 -655
  746. synth_ai/environments/examples/crafter_classic/agent_demos/old/trace_eval_OLD.py +0 -202
  747. synth_ai/environments/examples/crafter_classic/agent_demos/old/validate_openai_format.py +0 -166
  748. synth_ai/environments/examples/crafter_classic/config_logging.py +0 -111
  749. synth_ai/environments/examples/crafter_classic/debug_translation.py +0 -0
  750. synth_ai/environments/examples/crafter_classic/engine.py +0 -579
  751. synth_ai/environments/examples/crafter_classic/engine_deterministic_patch.py +0 -64
  752. synth_ai/environments/examples/crafter_classic/engine_helpers/action_map.py +0 -6
  753. synth_ai/environments/examples/crafter_classic/engine_helpers/serialization.py +0 -75
  754. synth_ai/environments/examples/crafter_classic/engine_serialization_patch_v3.py +0 -267
  755. synth_ai/environments/examples/crafter_classic/environment.py +0 -495
  756. synth_ai/environments/examples/crafter_classic/taskset.py +0 -233
  757. synth_ai/environments/examples/crafter_classic/trace_hooks_v3.py +0 -228
  758. synth_ai/environments/examples/crafter_classic/world_config_patch_simple.py +0 -299
  759. synth_ai/environments/examples/crafter_custom/__init__.py +0 -4
  760. synth_ai/environments/examples/crafter_custom/agent_demos/__init__.py +0 -1
  761. synth_ai/environments/examples/crafter_custom/agent_demos/trace_eval.py +0 -202
  762. synth_ai/environments/examples/crafter_custom/crafter/__init__.py +0 -7
  763. synth_ai/environments/examples/crafter_custom/crafter/config.py +0 -182
  764. synth_ai/environments/examples/crafter_custom/crafter/constants.py +0 -8
  765. synth_ai/environments/examples/crafter_custom/crafter/engine.py +0 -269
  766. synth_ai/environments/examples/crafter_custom/crafter/env.py +0 -262
  767. synth_ai/environments/examples/crafter_custom/crafter/objects.py +0 -417
  768. synth_ai/environments/examples/crafter_custom/crafter/recorder.py +0 -187
  769. synth_ai/environments/examples/crafter_custom/crafter/worldgen.py +0 -118
  770. synth_ai/environments/examples/crafter_custom/dataset_builder.py +0 -373
  771. synth_ai/environments/examples/crafter_custom/environment.py +0 -312
  772. synth_ai/environments/examples/crafter_custom/old/analyze_diamond_issue.py +0 -159
  773. synth_ai/environments/examples/crafter_custom/old/analyze_diamond_spawning.py +0 -158
  774. synth_ai/environments/examples/crafter_custom/old/compare_worlds.py +0 -71
  775. synth_ai/environments/examples/crafter_custom/old/dataset_stats.py +0 -105
  776. synth_ai/environments/examples/crafter_custom/old/diamond_spawning_summary.py +0 -119
  777. synth_ai/environments/examples/crafter_custom/old/example_dataset_usage.py +0 -52
  778. synth_ai/environments/examples/crafter_custom/run_dataset.py +0 -305
  779. synth_ai/environments/examples/enron/art_helpers/email_search_tools.py +0 -156
  780. synth_ai/environments/examples/enron/art_helpers/local_email_db.py +0 -281
  781. synth_ai/environments/examples/enron/art_helpers/types_enron.py +0 -25
  782. synth_ai/environments/examples/enron/engine.py +0 -300
  783. synth_ai/environments/examples/enron/environment.py +0 -234
  784. synth_ai/environments/examples/enron/taskset.py +0 -112
  785. synth_ai/environments/examples/enron/units/keyword_stats.py +0 -112
  786. synth_ai/environments/examples/minigrid/__init__.py +0 -48
  787. synth_ai/environments/examples/minigrid/agent_demos/minigrid_evaluation_framework.py +0 -1188
  788. synth_ai/environments/examples/minigrid/agent_demos/minigrid_quick_evaluation.py +0 -48
  789. synth_ai/environments/examples/minigrid/agent_demos/minigrid_react_agent.py +0 -562
  790. synth_ai/environments/examples/minigrid/agent_demos/minigrid_trace_evaluation.py +0 -221
  791. synth_ai/environments/examples/minigrid/engine.py +0 -589
  792. synth_ai/environments/examples/minigrid/environment.py +0 -274
  793. synth_ai/environments/examples/minigrid/environment_mapping.py +0 -242
  794. synth_ai/environments/examples/minigrid/puzzle_loader.py +0 -417
  795. synth_ai/environments/examples/minigrid/taskset.py +0 -583
  796. synth_ai/environments/examples/nethack/__init__.py +0 -7
  797. synth_ai/environments/examples/nethack/achievements.py +0 -337
  798. synth_ai/environments/examples/nethack/agent_demos/nethack_evaluation_framework.py +0 -981
  799. synth_ai/environments/examples/nethack/agent_demos/nethack_quick_evaluation.py +0 -74
  800. synth_ai/environments/examples/nethack/agent_demos/nethack_react_agent.py +0 -831
  801. synth_ai/environments/examples/nethack/engine.py +0 -739
  802. synth_ai/environments/examples/nethack/environment.py +0 -256
  803. synth_ai/environments/examples/nethack/helpers/__init__.py +0 -41
  804. synth_ai/environments/examples/nethack/helpers/action_mapping.py +0 -301
  805. synth_ai/environments/examples/nethack/helpers/nle_wrapper.py +0 -402
  806. synth_ai/environments/examples/nethack/helpers/observation_utils.py +0 -433
  807. synth_ai/environments/examples/nethack/helpers/recording_wrapper.py +0 -200
  808. synth_ai/environments/examples/nethack/helpers/trajectory_recorder.py +0 -269
  809. synth_ai/environments/examples/nethack/helpers/visualization/replay_viewer.py +0 -308
  810. synth_ai/environments/examples/nethack/helpers/visualization/visualizer.py +0 -431
  811. synth_ai/environments/examples/nethack/taskset.py +0 -323
  812. synth_ai/environments/examples/red/__init__.py +0 -7
  813. synth_ai/environments/examples/red/agent_demos/__init__.py +0 -1
  814. synth_ai/environments/examples/red/config_logging.py +0 -110
  815. synth_ai/environments/examples/red/engine.py +0 -721
  816. synth_ai/environments/examples/red/engine_helpers/__init__.py +0 -1
  817. synth_ai/environments/examples/red/engine_helpers/memory_map.py +0 -35
  818. synth_ai/environments/examples/red/engine_helpers/reward_components.py +0 -276
  819. synth_ai/environments/examples/red/engine_helpers/reward_library/__init__.py +0 -142
  820. synth_ai/environments/examples/red/engine_helpers/reward_library/adaptive_rewards.py +0 -57
  821. synth_ai/environments/examples/red/engine_helpers/reward_library/battle_rewards.py +0 -284
  822. synth_ai/environments/examples/red/engine_helpers/reward_library/composite_rewards.py +0 -150
  823. synth_ai/environments/examples/red/engine_helpers/reward_library/economy_rewards.py +0 -138
  824. synth_ai/environments/examples/red/engine_helpers/reward_library/efficiency_rewards.py +0 -57
  825. synth_ai/environments/examples/red/engine_helpers/reward_library/exploration_rewards.py +0 -331
  826. synth_ai/environments/examples/red/engine_helpers/reward_library/novelty_rewards.py +0 -121
  827. synth_ai/environments/examples/red/engine_helpers/reward_library/pallet_town_progression.py +0 -477
  828. synth_ai/environments/examples/red/engine_helpers/reward_library/pallet_town_rewards.py +0 -559
  829. synth_ai/environments/examples/red/engine_helpers/reward_library/pokemon_rewards.py +0 -313
  830. synth_ai/environments/examples/red/engine_helpers/reward_library/social_rewards.py +0 -148
  831. synth_ai/environments/examples/red/engine_helpers/reward_library/story_rewards.py +0 -247
  832. synth_ai/environments/examples/red/engine_helpers/screen_analysis.py +0 -368
  833. synth_ai/environments/examples/red/engine_helpers/state_extraction.py +0 -172
  834. synth_ai/environments/examples/red/environment.py +0 -298
  835. synth_ai/environments/examples/red/taskset.py +0 -79
  836. synth_ai/environments/examples/red/units/__init__.py +0 -1
  837. synth_ai/environments/examples/sokoban/__init__.py +0 -1
  838. synth_ai/environments/examples/sokoban/agent_demos/sokoban_full_eval.py +0 -899
  839. synth_ai/environments/examples/sokoban/engine.py +0 -678
  840. synth_ai/environments/examples/sokoban/engine_helpers/__init__.py +0 -1
  841. synth_ai/environments/examples/sokoban/engine_helpers/room_utils.py +0 -657
  842. synth_ai/environments/examples/sokoban/engine_helpers/vendored/__init__.py +0 -18
  843. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/__init__.py +0 -3
  844. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/boxoban_env.py +0 -131
  845. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/render_utils.py +0 -370
  846. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/room_utils.py +0 -332
  847. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env.py +0 -306
  848. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env_fixed_targets.py +0 -67
  849. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env_pull.py +0 -115
  850. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env_two_player.py +0 -123
  851. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env_variations.py +0 -394
  852. synth_ai/environments/examples/sokoban/environment.py +0 -229
  853. synth_ai/environments/examples/sokoban/generate_verified_puzzles.py +0 -440
  854. synth_ai/environments/examples/sokoban/puzzle_loader.py +0 -312
  855. synth_ai/environments/examples/sokoban/taskset.py +0 -544
  856. synth_ai/environments/examples/tictactoe/__init__.py +0 -1
  857. synth_ai/environments/examples/tictactoe/engine.py +0 -368
  858. synth_ai/environments/examples/tictactoe/environment.py +0 -240
  859. synth_ai/environments/examples/tictactoe/taskset.py +0 -215
  860. synth_ai/environments/examples/verilog/__init__.py +0 -10
  861. synth_ai/environments/examples/verilog/engine.py +0 -421
  862. synth_ai/environments/examples/verilog/environment.py +0 -350
  863. synth_ai/environments/examples/verilog/taskset.py +0 -420
  864. synth_ai/environments/examples/wordle/__init__.py +0 -29
  865. synth_ai/environments/examples/wordle/engine.py +0 -398
  866. synth_ai/environments/examples/wordle/environment.py +0 -159
  867. synth_ai/environments/examples/wordle/helpers/generate_instances_wordfreq.py +0 -75
  868. synth_ai/environments/examples/wordle/taskset.py +0 -230
  869. synth_ai/environments/reproducibility/core.py +0 -42
  870. synth_ai/environments/reproducibility/helpers.py +0 -0
  871. synth_ai/environments/reproducibility/tree.py +0 -363
  872. synth_ai/environments/service/app.py +0 -97
  873. synth_ai/environments/service/core_routes.py +0 -1021
  874. synth_ai/environments/service/external_registry.py +0 -56
  875. synth_ai/environments/service/registry.py +0 -9
  876. synth_ai/environments/stateful/__init__.py +0 -1
  877. synth_ai/environments/stateful/core.py +0 -163
  878. synth_ai/environments/stateful/engine.py +0 -21
  879. synth_ai/environments/stateful/state.py +0 -7
  880. synth_ai/environments/tasks/api.py +0 -19
  881. synth_ai/environments/tasks/core.py +0 -81
  882. synth_ai/environments/tasks/filters.py +0 -40
  883. synth_ai/environments/tasks/utils.py +0 -90
  884. synth_ai/environments/v0_observability/history.py +0 -3
  885. synth_ai/environments/v0_observability/log.py +0 -2
  886. synth_ai/evals/__init__.py +0 -15
  887. synth_ai/evals/base.py +0 -13
  888. synth_ai/evals/client.py +0 -82
  889. synth_ai/handshake.py +0 -109
  890. synth_ai/http.py +0 -26
  891. synth_ai/http_client.py +0 -136
  892. synth_ai/inference/__init__.py +0 -5
  893. synth_ai/inference/client.py +0 -34
  894. synth_ai/jobs/client.py +0 -295
  895. synth_ai/judge_schemas.py +0 -127
  896. synth_ai/learning/__init__.py +0 -59
  897. synth_ai/learning/client.py +0 -241
  898. synth_ai/learning/ft_client.py +0 -7
  899. synth_ai/learning/health.py +0 -49
  900. synth_ai/learning/jobs.py +0 -201
  901. synth_ai/learning/rl/client.py +0 -267
  902. synth_ai/learning/rl/contracts.py +0 -27
  903. synth_ai/learning/rl/env_keys.py +0 -166
  904. synth_ai/learning/rl/secrets.py +0 -13
  905. synth_ai/learning/sft/client.py +0 -68
  906. synth_ai/learning/sft/config.py +0 -270
  907. synth_ai/learning/sft/data.py +0 -295
  908. synth_ai/learning/validators.py +0 -49
  909. synth_ai/lm/__init__.py +0 -25
  910. synth_ai/task/__init__.py +0 -121
  911. synth_ai/task/apps/__init__.py +0 -129
  912. synth_ai/task/config.py +0 -257
  913. synth_ai/task/contracts.py +0 -236
  914. synth_ai/task/datasets.py +0 -108
  915. synth_ai/task/proxy.py +0 -251
  916. synth_ai/task/rubrics/__init__.py +0 -56
  917. synth_ai/task/rubrics/loaders.py +0 -152
  918. synth_ai/task/server.py +0 -432
  919. synth_ai/task/trace_correlation_helpers.py +0 -315
  920. synth_ai/task/tracing_utils.py +0 -84
  921. synth_ai/task/validators.py +0 -418
  922. synth_ai/tracing_v3/__init__.py +0 -97
  923. synth_ai/tracing_v3/abstractions.py +0 -302
  924. synth_ai/tracing_v3/config.py +0 -84
  925. synth_ai/tracing_v3/db_config.py +0 -194
  926. synth_ai/tracing_v3/decorators.py +0 -398
  927. synth_ai/tracing_v3/llm_call_record_helpers.py +0 -391
  928. synth_ai/tracing_v3/migration_helper.py +0 -120
  929. synth_ai/tracing_v3/session_tracer.py +0 -540
  930. synth_ai/tracing_v3/storage/base.py +0 -210
  931. synth_ai/tracing_v3/storage/config.py +0 -75
  932. synth_ai/tracing_v3/storage/factory.py +0 -39
  933. synth_ai/tracing_v3/trace_utils.py +0 -317
  934. synth_ai/tracing_v3/turso/daemon.py +0 -151
  935. synth_ai/tracing_v3/turso/models.py +0 -469
  936. synth_ai/tracing_v3/turso/native_manager.py +0 -1209
  937. synth_ai/tracing_v3/utils.py +0 -108
  938. synth_ai/tui/__init__.py +0 -5
  939. synth_ai/tui/__main__.py +0 -13
  940. synth_ai/tui/cli/__init__.py +0 -1
  941. synth_ai/tui/cli/query_experiments.py +0 -164
  942. synth_ai/tui/cli/query_experiments_v3.py +0 -164
  943. synth_ai/tui/dashboard.py +0 -906
  944. synth_ai/v0/api/__init__.py +0 -8
  945. synth_ai/v0/api/models/__init__.py +0 -8
  946. synth_ai/v0/api/models/supported.py +0 -8
  947. synth_ai/v0/config/__init__.py +0 -15
  948. synth_ai/v0/config/base_url.py +0 -12
  949. synth_ai/v0/lm/__init__.py +0 -51
  950. synth_ai/v0/lm/caching/__init__.py +0 -0
  951. synth_ai/v0/lm/caching/constants.py +0 -6
  952. synth_ai/v0/lm/caching/dbs.py +0 -0
  953. synth_ai/v0/lm/caching/ephemeral.py +0 -100
  954. synth_ai/v0/lm/caching/handler.py +0 -137
  955. synth_ai/v0/lm/caching/initialize.py +0 -11
  956. synth_ai/v0/lm/caching/persistent.py +0 -114
  957. synth_ai/v0/lm/config.py +0 -115
  958. synth_ai/v0/lm/constants.py +0 -32
  959. synth_ai/v0/lm/core/__init__.py +0 -8
  960. synth_ai/v0/lm/core/all.py +0 -73
  961. synth_ai/v0/lm/core/exceptions.py +0 -5
  962. synth_ai/v0/lm/core/main.py +0 -331
  963. synth_ai/v0/lm/core/main_v3.py +0 -594
  964. synth_ai/v0/lm/core/synth_models.py +0 -35
  965. synth_ai/v0/lm/core/vendor_clients.py +0 -190
  966. synth_ai/v0/lm/cost/__init__.py +0 -0
  967. synth_ai/v0/lm/cost/monitor.py +0 -1
  968. synth_ai/v0/lm/cost/statefulness.py +0 -1
  969. synth_ai/v0/lm/injection.py +0 -80
  970. synth_ai/v0/lm/overrides.py +0 -206
  971. synth_ai/v0/lm/provider_support/__init__.py +0 -8
  972. synth_ai/v0/lm/provider_support/anthropic.py +0 -972
  973. synth_ai/v0/lm/provider_support/openai.py +0 -1139
  974. synth_ai/v0/lm/provider_support/suppress_logging.py +0 -31
  975. synth_ai/v0/lm/structured_outputs/__init__.py +0 -0
  976. synth_ai/v0/lm/structured_outputs/handler.py +0 -440
  977. synth_ai/v0/lm/structured_outputs/inject.py +0 -297
  978. synth_ai/v0/lm/structured_outputs/rehabilitate.py +0 -185
  979. synth_ai/v0/lm/tools/__init__.py +0 -3
  980. synth_ai/v0/lm/tools/base.py +0 -172
  981. synth_ai/v0/lm/unified_interface.py +0 -202
  982. synth_ai/v0/lm/vendors/__init__.py +0 -0
  983. synth_ai/v0/lm/vendors/base.py +0 -81
  984. synth_ai/v0/lm/vendors/core/__init__.py +0 -0
  985. synth_ai/v0/lm/vendors/core/anthropic_api.py +0 -387
  986. synth_ai/v0/lm/vendors/core/gemini_api.py +0 -292
  987. synth_ai/v0/lm/vendors/core/mistral_api.py +0 -322
  988. synth_ai/v0/lm/vendors/core/openai_api.py +0 -227
  989. synth_ai/v0/lm/vendors/core/synth_dev_api.py +0 -0
  990. synth_ai/v0/lm/vendors/local/__init__.py +0 -0
  991. synth_ai/v0/lm/vendors/local/ollama.py +0 -0
  992. synth_ai/v0/lm/vendors/openai_standard.py +0 -782
  993. synth_ai/v0/lm/vendors/openai_standard_responses.py +0 -259
  994. synth_ai/v0/lm/vendors/retries.py +0 -22
  995. synth_ai/v0/lm/vendors/supported/__init__.py +0 -0
  996. synth_ai/v0/lm/vendors/supported/custom_endpoint.py +0 -415
  997. synth_ai/v0/lm/vendors/supported/deepseek.py +0 -69
  998. synth_ai/v0/lm/vendors/supported/grok.py +0 -75
  999. synth_ai/v0/lm/vendors/supported/groq.py +0 -16
  1000. synth_ai/v0/lm/vendors/supported/ollama.py +0 -15
  1001. synth_ai/v0/lm/vendors/supported/openrouter.py +0 -74
  1002. synth_ai/v0/lm/vendors/supported/together.py +0 -11
  1003. synth_ai/v0/lm/vendors/synth_client.py +0 -835
  1004. synth_ai/v0/lm/warmup.py +0 -186
  1005. synth_ai/v0/tracing/__init__.py +0 -0
  1006. synth_ai/v0/tracing/abstractions.py +0 -224
  1007. synth_ai/v0/tracing/base_client.py +0 -91
  1008. synth_ai/v0/tracing/client_manager.py +0 -131
  1009. synth_ai/v0/tracing/config.py +0 -142
  1010. synth_ai/v0/tracing/context.py +0 -146
  1011. synth_ai/v0/tracing/decorators.py +0 -682
  1012. synth_ai/v0/tracing/events/__init__.py +0 -0
  1013. synth_ai/v0/tracing/events/manage.py +0 -147
  1014. synth_ai/v0/tracing/events/scope.py +0 -86
  1015. synth_ai/v0/tracing/events/store.py +0 -228
  1016. synth_ai/v0/tracing/immediate_client.py +0 -151
  1017. synth_ai/v0/tracing/local.py +0 -18
  1018. synth_ai/v0/tracing/log_client_base.py +0 -73
  1019. synth_ai/v0/tracing/retry_queue.py +0 -186
  1020. synth_ai/v0/tracing/trackers.py +0 -515
  1021. synth_ai/v0/tracing/upload.py +0 -409
  1022. synth_ai/v0/tracing/utils.py +0 -9
  1023. synth_ai/v0/tracing_v1/__init__.py +0 -16
  1024. synth_ai/v0/tracing_v1/abstractions.py +0 -224
  1025. synth_ai/v0/tracing_v1/base_client.py +0 -91
  1026. synth_ai/v0/tracing_v1/client_manager.py +0 -131
  1027. synth_ai/v0/tracing_v1/config.py +0 -142
  1028. synth_ai/v0/tracing_v1/context.py +0 -146
  1029. synth_ai/v0/tracing_v1/decorators.py +0 -703
  1030. synth_ai/v0/tracing_v1/events/__init__.py +0 -0
  1031. synth_ai/v0/tracing_v1/events/manage.py +0 -147
  1032. synth_ai/v0/tracing_v1/events/scope.py +0 -86
  1033. synth_ai/v0/tracing_v1/events/store.py +0 -228
  1034. synth_ai/v0/tracing_v1/immediate_client.py +0 -151
  1035. synth_ai/v0/tracing_v1/local.py +0 -18
  1036. synth_ai/v0/tracing_v1/log_client_base.py +0 -73
  1037. synth_ai/v0/tracing_v1/retry_queue.py +0 -186
  1038. synth_ai/v0/tracing_v1/trackers.py +0 -515
  1039. synth_ai/v0/tracing_v1/upload.py +0 -527
  1040. synth_ai/v0/tracing_v1/utils.py +0 -9
  1041. synth_ai/v0/tracing_v3/__init__.py +0 -10
  1042. synth_ai/v0/tracing_v3/abstractions.py +0 -3
  1043. synth_ai/v0/tracing_v3/decorators.py +0 -3
  1044. synth_ai/v0/tracing_v3/llm_call_record_helpers.py +0 -3
  1045. synth_ai/v0/tracing_v3/session_tracer.py +0 -3
  1046. synth_ai-0.2.14.dist-info/METADATA +0 -139
  1047. synth_ai-0.2.14.dist-info/RECORD +0 -762
  1048. synth_ai-0.2.14.dist-info/top_level.txt +0 -2
  1049. /synth_ai/{demos/demo_task_apps → cli/demo_apps}/crafter/__init__.py +0 -0
  1050. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/__init__.py +0 -0
  1051. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/crafter/configs/crafter_fft_4b.toml +0 -0
  1052. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/crafter/configs/rl_from_base_qwen4b.toml +0 -0
  1053. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/math/__init__.py +0 -0
  1054. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/math/_common.py +0 -0
  1055. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/math/app.py +0 -0
  1056. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/math/config.toml +0 -0
  1057. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/math/deploy_modal.py +0 -0
  1058. {examples/task_apps → synth_ai/core/apps}/__init__.py +0 -0
  1059. /synth_ai/{tracing_v3 → core/tracing_v3}/examples/basic_usage.py +0 -0
  1060. /synth_ai/{tracing_v3 → core/tracing_v3}/hooks.py +0 -0
  1061. /synth_ai/{tracing_v3 → core/tracing_v3}/lm_call_record_abstractions.py +0 -0
  1062. /synth_ai/{tracing_v3 → core/tracing_v3}/replica_sync.py +0 -0
  1063. /synth_ai/{tracing_v3 → core/tracing_v3}/serialization.py +0 -0
  1064. /synth_ai/{tracing_v3 → core/tracing_v3}/storage/__init__.py +0 -0
  1065. /synth_ai/{tracing_v3 → core/tracing_v3}/storage/exceptions.py +0 -0
  1066. /synth_ai/{tracing_v3 → core/tracing_v3}/storage/types.py +0 -0
  1067. /synth_ai/{tracing_v3 → core/tracing_v3}/storage/utils.py +0 -0
  1068. /synth_ai/{tracing_v3 → core/tracing_v3}/turso/__init__.py +0 -0
  1069. /synth_ai/{evals → sdk/judging}/types.py +0 -0
  1070. /synth_ai/{learning → sdk/learning}/algorithms.py +0 -0
  1071. /synth_ai/{learning → sdk/learning}/config.py +0 -0
  1072. /synth_ai/{learning → sdk/learning}/constants.py +0 -0
  1073. /synth_ai/{learning → sdk/learning}/core.py +0 -0
  1074. /synth_ai/{learning → sdk/learning}/gateway.py +0 -0
  1075. /synth_ai/{learning → sdk/learning}/rl/__init__.py +0 -0
  1076. /synth_ai/{learning → sdk/learning}/rl/config.py +0 -0
  1077. /synth_ai/{learning → sdk/learning}/rl_client.py +0 -0
  1078. /synth_ai/{learning → sdk/learning}/sft/__init__.py +0 -0
  1079. /synth_ai/{learning → sdk/learning}/sse.py +0 -0
  1080. /synth_ai/{task → sdk/task}/auth.py +0 -0
  1081. /synth_ai/{task → sdk/task}/client.py +0 -0
  1082. /synth_ai/{task → sdk/task}/errors.py +0 -0
  1083. /synth_ai/{task → sdk/task}/health.py +0 -0
  1084. /synth_ai/{task → sdk/task}/json.py +0 -0
  1085. /synth_ai/{task → sdk/task}/rubrics/models.py +0 -0
  1086. /synth_ai/{task → sdk/task}/rubrics/scoring.py +0 -0
  1087. /synth_ai/{task → sdk/task}/rubrics/strict.py +0 -0
  1088. /synth_ai/{task → sdk/task}/vendors.py +0 -0
  1089. {synth_ai-0.2.14.dist-info → synth_ai-0.4.1.dist-info}/WHEEL +0 -0
  1090. {synth_ai-0.2.14.dist-info → synth_ai-0.4.1.dist-info}/entry_points.txt +0 -0
  1091. {synth_ai-0.2.14.dist-info → synth_ai-0.4.1.dist-info}/licenses/LICENSE +0 -0
@@ -1,495 +0,0 @@
1
- """CrafterClassicEnvironment — thin wrapper exposing CrafterEngine via StatefulEnvironment API."""
2
-
3
- from __future__ import annotations
4
-
5
- import base64
6
- import dataclasses
7
- import logging
8
- import time
9
- from io import BytesIO
10
- from typing import Any, Dict, List, Optional, Union
11
-
12
- import numpy as np
13
- from PIL import Image
14
-
15
- # Import tracing abstractions
16
- from synth_ai.tracing_v3.abstractions import (
17
- RuntimeEvent,
18
- SessionEventMarkovBlanketMessage,
19
- TimeRecord,
20
- )
21
-
22
- # Import logging configuration to suppress JAX debug messages
23
- from .config_logging import safe_compare
24
-
25
- logger = logging.getLogger(__name__)
26
-
27
- from pydantic import BaseModel, Field
28
-
29
- from synth_ai.environments.environment.shared_engine import (
30
- GetObservationCallable,
31
- InternalObservation,
32
- )
33
- from synth_ai.environments.environment.tools import (
34
- TOOL_REGISTRY,
35
- AbstractTool,
36
- EnvToolCall,
37
- ToolResult,
38
- register_tool,
39
- )
40
- from synth_ai.environments.examples.crafter_classic.engine import (
41
- CrafterEngine,
42
- CrafterEngineSnapshot,
43
- CrafterPrivateState,
44
- CrafterPublicState,
45
- )
46
- from synth_ai.environments.examples.crafter_classic.taskset import CrafterTaskInstance
47
- from synth_ai.environments.reproducibility.core import ReproducibleEnvironment
48
- from synth_ai.environments.stateful.core import StatefulEnvironment
49
-
50
-
51
- def _convert_numpy_to_python(obj: Any) -> Any:
52
- if isinstance(obj, np.integer):
53
- return int(obj)
54
- if isinstance(obj, np.floating):
55
- return float(obj)
56
- if isinstance(obj, np.ndarray):
57
- return obj.tolist()
58
- if isinstance(obj, dict):
59
- return {k: _convert_numpy_to_python(v) for k, v in obj.items()}
60
- if isinstance(obj, (list, tuple)):
61
- return [_convert_numpy_to_python(item) for item in obj]
62
- return obj
63
-
64
-
65
- def _encode_image_to_base64(image_array: Any) -> dict[str, Any] | None:
66
- if not isinstance(image_array, np.ndarray):
67
- return None
68
- if image_array.ndim != 3 or image_array.shape[-1] not in (1, 3, 4):
69
- return None
70
- try:
71
- array_uint8 = (
72
- image_array.astype("uint8")
73
- if image_array.dtype != np.uint8
74
- else image_array # pragma: no cover - fast path
75
- )
76
- mode = "L" if array_uint8.shape[-1] == 1 else "RGB"
77
- if array_uint8.shape[-1] == 4:
78
- mode = "RGBA"
79
- img = Image.fromarray(array_uint8, mode=mode)
80
- buffer = BytesIO()
81
- img.save(buffer, format="PNG")
82
- encoded = base64.b64encode(buffer.getvalue()).decode("ascii")
83
- width = int(array_uint8.shape[1])
84
- height = int(array_uint8.shape[0])
85
- return {
86
- "format": "png",
87
- "width": width,
88
- "height": height,
89
- "data": encoded,
90
- "data_url": f"data:image/png;base64,{encoded}",
91
- }
92
- except Exception:
93
- return None
94
-
95
-
96
- # --- Tool Definition ---
97
- class CrafterActionInput(BaseModel):
98
- action: int = Field(..., description="Integer action for the Crafter environment.")
99
-
100
-
101
- class CrafterInteractTool(AbstractTool):
102
- name = "interact"
103
- description = "Performs an action in the Crafter environment."
104
- call_schema = CrafterActionInput
105
- result_schema = ToolResult
106
-
107
- def __init__(self, engine: CrafterEngine, session_tracer: Optional[Any] = None):
108
- self.engine = engine
109
- self.session_tracer = session_tracer
110
-
111
- async def __call__(self, call: EnvToolCall) -> ToolResult:
112
- try:
113
- # Store state before execution
114
- state_before = {"action_args": call.args}
115
-
116
- validated_args = self.call_schema(**call.args)
117
- action_to_pass = self.engine._validate_action_engine(validated_args.action)
118
-
119
- # Execute the engine step
120
- priv_state, pub_state = await self.engine._step_engine(action_to_pass)
121
-
122
- # Store state after execution
123
- state_after = {
124
- "engine_result": {"private_state": priv_state, "public_state": pub_state}
125
- }
126
-
127
- # Record runtime event for tool execution
128
- if (
129
- self.session_tracer
130
- and hasattr(self.session_tracer, "current_session")
131
- and self.session_tracer.current_session
132
- ):
133
- runtime_execution_event = RuntimeEvent()
134
- runtime_execution_event.time_record = TimeRecord()
135
- runtime_execution_event.time_record.event_time = time.time()
136
- runtime_execution_event.time_record.message_time = None
137
- runtime_execution_event.system_instance_id = "crafter_interact_tool"
138
- runtime_execution_event.system_state_before = state_before
139
- runtime_execution_event.system_state_after = state_after
140
- runtime_execution_event.actions = [action_to_pass]
141
- runtime_execution_event.metadata = {"execution_step": "engine_action"}
142
- # Add directly to event history, bypassing timestep requirement
143
- self.session_tracer.current_session.add_event(runtime_execution_event)
144
-
145
- return ToolResult(
146
- ok=True,
147
- payload={
148
- "public_state": pub_state,
149
- "private_state": priv_state,
150
- },
151
- )
152
- except Exception as e:
153
- pub_state_on_error = self.engine._get_public_state_from_env() # Use engine helper
154
- # Get a safe private state for error cases
155
- health_dead = safe_compare(0, self.engine.env._player.health, ">=")
156
- step_exceeded = safe_compare(self.engine.env._length, self.engine.env._step, "<=")
157
- priv_state_on_error = self.engine._get_private_state_from_env(
158
- 0, health_dead, step_exceeded
159
- )
160
- return ToolResult(
161
- ok=False,
162
- error=str(e),
163
- payload={
164
- "public_state": pub_state_on_error,
165
- "private_state": priv_state_on_error,
166
- },
167
- )
168
-
169
-
170
- # Default observation callable (can be customized via __init__)
171
- class SynthCrafterObservationCallable(GetObservationCallable):
172
- """Default observation: public state dict + per-step reward/flags.
173
-
174
- Additionally computes a small local semantic patch centered on the player
175
- to simplify visualization on the client. The patch is exposed under the
176
- key `semantic_map_patch7` as a list-of-lists of ints (7x7 unless the
177
- semantic map is smaller, in which case it is cropped at edges).
178
- """
179
-
180
- def __init__(self, view_size: int = 7) -> None:
181
- self.view_size = max(1, int(view_size))
182
-
183
- async def get_observation(
184
- self, pub: CrafterPublicState, priv: CrafterPrivateState
185
- ) -> InternalObservation:
186
- obs_dict: Dict[str, Any] = dataclasses.asdict(pub) # type: ignore
187
- obs_dict["reward_last_step"] = priv.reward_last_step
188
- obs_dict["total_reward_episode"] = priv.total_reward_episode
189
- obs_dict["terminated"] = priv.terminated
190
- obs_dict["truncated"] = priv.truncated
191
- if pub.error_info:
192
- obs_dict["tool_error"] = pub.error_info
193
- counts_payload = {}
194
- try:
195
- counts = getattr(priv, "achievements_current_values", {}) or {}
196
- for k, v in counts.items():
197
- try:
198
- counts_payload[str(k)] = int(v)
199
- except Exception:
200
- try:
201
- counts_payload[str(k)] = int(float(v))
202
- except Exception:
203
- continue
204
- if counts_payload:
205
- obs_dict["achievements_counts"] = counts_payload
206
- except Exception:
207
- # Best effort; omit counts if coercion fails
208
- pass
209
-
210
- # Derive a simple local semantic patch around the player for easy rendering
211
- try:
212
- sem = pub.semantic_map
213
- if sem is not None:
214
- rows = int(getattr(sem, "shape", [0, 0])[0]) # type: ignore
215
- cols = int(getattr(sem, "shape", [0, 0])[1]) # type: ignore
216
- if rows > 0 and cols > 0:
217
- px, py = int(pub.player_position[0]), int(pub.player_position[1])
218
- half = max(1, self.view_size // 2)
219
- x0, y0 = px - half, py - half
220
- x1, y1 = px + half, py + half
221
- patch: list[list[int]] = []
222
- for gy in range(y0, y1 + 1):
223
- row_vals: list[int] = []
224
- for gx in range(x0, x1 + 1):
225
- if 0 <= gy < rows and 0 <= gx < cols:
226
- try:
227
- val = int(sem[gy, gx]) # type: ignore[index]
228
- except Exception:
229
- val = 0
230
- else:
231
- val = 0
232
- row_vals.append(val)
233
- patch.append(row_vals)
234
- obs_dict["semantic_map_patch7"] = patch
235
- except Exception:
236
- # Best-effort; omit patch on error
237
- pass
238
-
239
- return obs_dict
240
-
241
-
242
- class CrafterClassicEnvironment(StatefulEnvironment, ReproducibleEnvironment[CrafterEngine]):
243
- """Environment wrapper bridging agent tool‑calls to `crafter.Env` dynamics."""
244
-
245
- def __init__(
246
- self,
247
- task_instance: "CrafterTaskInstance",
248
- custom_step_obs: Optional[GetObservationCallable] = None,
249
- custom_ckpt_obs: Optional[GetObservationCallable] = None,
250
- session_tracer: Optional[Any] = None, # SessionTracer from higher level
251
- ) -> None:
252
- self.name = "CrafterClassic"
253
- self.task_instance = task_instance
254
- self.custom_step_observation_callable = custom_step_obs or SynthCrafterObservationCallable()
255
- self.custom_checkpoint_observation_callable = (
256
- custom_ckpt_obs or SynthCrafterObservationCallable()
257
- )
258
- self.engine = CrafterEngine(task_instance)
259
- self.session_tracer = session_tracer # Store tracer for runtime events
260
-
261
- self._interact_tool = CrafterInteractTool(self.engine, session_tracer=session_tracer)
262
- if self._interact_tool.name not in TOOL_REGISTRY:
263
- register_tool(self._interact_tool)
264
-
265
- # ────────────────────────────────────────────────────────────────────
266
- # Lifecycle helpers
267
- # ────────────────────────────────────────────────────────────────────
268
-
269
- async def initialize(self, seed: Optional[int] = None) -> InternalObservation: # type: ignore[override]
270
- # Check if seed was provided in task instance metadata
271
- if (
272
- seed is None
273
- and hasattr(self.task_instance, "metadata")
274
- and hasattr(self.task_instance.metadata, "seed")
275
- ):
276
- seed = self.task_instance.metadata.seed
277
- # Check if seed was provided in initial_engine_snapshot
278
- elif (
279
- seed is None
280
- and hasattr(self.task_instance, "initial_engine_snapshot")
281
- and isinstance(self.task_instance.initial_engine_snapshot, dict)
282
- ):
283
- seed = self.task_instance.initial_engine_snapshot.get("seed")
284
-
285
- # Initialize with seed from various sources
286
-
287
- priv, pub = await self.engine._reset_engine(seed=seed)
288
- return await self._to_observation(priv, pub, self.custom_step_observation_callable)
289
-
290
- async def terminate(self) -> InternalObservation: # type: ignore[override]
291
- pub = self.engine._get_public_state_from_env()
292
- priv = self.engine._get_private_state_from_env(0, True, False) # Terminated state
293
- priv.terminated = True
294
- obs_dict = {"status": "Environment terminated."}
295
- return await self._to_observation(
296
- priv, pub, self.custom_step_observation_callable, extra_obs=obs_dict
297
- )
298
-
299
- # ────────────────────────────────────────────────────────────────────
300
- # Step + checkpoint
301
- # ────────────────────────────────────────────────────────────────────
302
-
303
- def validate_tool_calls(
304
- self, tool_calls: Union[EnvToolCall, List[EnvToolCall], List[List[EnvToolCall]]]
305
- ) -> EnvToolCall:
306
- # Store the original tool calls for tracing
307
- state_before = {"tool_calls": tool_calls}
308
-
309
- # Normalize and validate to a single EnvToolCall (same as Sokoban)
310
- if isinstance(tool_calls, list):
311
- if not tool_calls:
312
- raise ValueError("Received empty list of tool calls.")
313
- if isinstance(tool_calls[0], list):
314
- if not tool_calls[0]:
315
- raise ValueError("Received empty inner list of tool calls.")
316
- agent_call = tool_calls[0][0]
317
- else:
318
- agent_call = tool_calls[0]
319
- elif isinstance(tool_calls, EnvToolCall):
320
- agent_call = tool_calls
321
- else:
322
- raise TypeError(f"Unexpected type for tool_calls: {type(tool_calls)}")
323
-
324
- if not isinstance(agent_call, EnvToolCall):
325
- raise TypeError(f"Processed call is not EnvToolCall: {type(agent_call)}")
326
- if agent_call.tool != "interact":
327
- raise ValueError(f"Unknown tool: {agent_call.tool}. Expected 'interact'.")
328
-
329
- # Record runtime event for tool call validation
330
- if (
331
- self.session_tracer
332
- and hasattr(self.session_tracer, "current_session")
333
- and self.session_tracer.current_session
334
- ):
335
- runtime_validation_event = RuntimeEvent()
336
- runtime_validation_event.time_record = TimeRecord()
337
- runtime_validation_event.time_record.event_time = time.time()
338
- runtime_validation_event.time_record.message_time = None
339
- runtime_validation_event.system_instance_id = "crafter_environment"
340
- runtime_validation_event.system_state_before = state_before
341
- runtime_validation_event.system_state_after = {"validated_call": agent_call}
342
- runtime_validation_event.metadata = {"validation_step": "tool_call_validation"}
343
- # Add directly to event history, bypassing timestep requirement
344
- self.session_tracer.current_session.add_event(runtime_validation_event)
345
-
346
- return agent_call
347
-
348
- async def step(
349
- self, tool_calls: Union[EnvToolCall, List[EnvToolCall], List[List[EnvToolCall]]]
350
- ) -> InternalObservation: # type: ignore[override]
351
- step_start_time = time.time()
352
- agent_call = self.validate_tool_calls(tool_calls)
353
- interact_start = time.time()
354
- tool_result: ToolResult = await self._interact_tool(agent_call)
355
- interact_time = time.time() - interact_start
356
-
357
- payload_dict = tool_result.payload
358
- pub_state: CrafterPublicState
359
- priv_state: CrafterPrivateState
360
-
361
- if tool_result.ok:
362
- # payload contains the actual state objects from the interact tool
363
- priv_state = payload_dict.get("private_state")
364
- pub_state = payload_dict.get("public_state")
365
-
366
- # Validate we got the expected state objects
367
- if not isinstance(priv_state, CrafterPrivateState) or not isinstance(
368
- pub_state, CrafterPublicState
369
- ):
370
- logger.error(
371
- f"Invalid state types in payload: priv={type(priv_state)}, pub={type(pub_state)}"
372
- )
373
- # Fall back to getting current state
374
- pub_state = self.engine._get_public_state_from_env()
375
- health_dead = safe_compare(0, self.engine.env._player.health, ">=")
376
- step_exceeded = safe_compare(self.engine.env._length, self.engine.env._step, "<=")
377
- priv_state = self.engine._get_private_state_from_env(0, health_dead, step_exceeded)
378
- pub_state.error_info = "Invalid state types in tool result"
379
- else:
380
- # Tool call failed, use states from payload if available, otherwise get current state
381
- priv_state = payload_dict.get("private_state")
382
- pub_state = payload_dict.get("public_state")
383
-
384
- if not isinstance(priv_state, CrafterPrivateState) or not isinstance(
385
- pub_state, CrafterPublicState
386
- ):
387
- # Fall back to getting current state
388
- pub_state = self.engine._get_public_state_from_env()
389
- health_dead = safe_compare(0, self.engine.env._player.health, ">=")
390
- step_exceeded = safe_compare(self.engine.env._length, self.engine.env._step, "<=")
391
- priv_state = self.engine._get_private_state_from_env(0, health_dead, step_exceeded)
392
-
393
- if tool_result.error:
394
- pub_state.error_info = tool_result.error
395
-
396
- obs = await self._to_observation(
397
- priv_state, pub_state, self.custom_step_observation_callable
398
- )
399
- total_step_time = time.time() - step_start_time
400
- logger.debug(
401
- f"CrafterClassic step completed in {total_step_time:.3f}s (interact: {interact_time:.3f}s)"
402
- )
403
- return obs
404
-
405
- async def checkpoint(self) -> InternalObservation: # type: ignore[override]
406
- engine_snapshot: CrafterEngineSnapshot = await self.engine._serialize_engine()
407
- priv = self.engine._get_private_state_from_env(0, False, False) # Get current state for obs
408
- pub = self.engine._get_public_state_from_env()
409
- obs_data = await self._to_observation(
410
- priv, pub, self.custom_checkpoint_observation_callable
411
- )
412
- if isinstance(obs_data, dict):
413
- obs_data["engine_snapshot_data"] = engine_snapshot.model_dump()
414
- return obs_data
415
-
416
- # ────────────────────────────────────────────────────────────────────
417
- # Helpers
418
- # ────────────────────────────────────────────────────────────────────
419
-
420
- async def _to_observation(
421
- self,
422
- priv: CrafterPrivateState,
423
- pub: CrafterPublicState,
424
- obs_cb: Optional[GetObservationCallable],
425
- extra_obs: Optional[Dict[str, Any]] = None,
426
- ) -> InternalObservation:
427
- # Store state before observation generation
428
- state_before = {"private_state": priv, "public_state": pub}
429
-
430
- active_obs_cb = obs_cb or SynthCrafterObservationCallable()
431
- raw_observation = await active_obs_cb.get_observation(pub, priv)
432
- observation = self._prepare_observation(raw_observation)
433
- if extra_obs and isinstance(observation, dict):
434
- observation.update(extra_obs)
435
-
436
- # Record runtime event for observation generation
437
- if (
438
- self.session_tracer
439
- and hasattr(self.session_tracer, "current_session")
440
- and self.session_tracer.current_session
441
- ):
442
- runtime_obs_event = RuntimeEvent()
443
- runtime_obs_event.time_record = TimeRecord()
444
- runtime_obs_event.time_record.event_time = time.time()
445
- runtime_obs_event.time_record.message_time = None
446
- runtime_obs_event.system_instance_id = "observation_generator"
447
- runtime_obs_event.system_state_before = state_before
448
- runtime_obs_event.system_state_after = {"observation": observation}
449
- runtime_obs_event.metadata = {"observation_step": "state_to_obs_conversion"}
450
- # Add directly to event history, bypassing timestep requirement
451
- self.session_tracer.current_session.add_event(runtime_obs_event)
452
-
453
- return observation
454
-
455
- def _prepare_observation(self, observation: Any) -> dict[str, Any]:
456
- obs_dict: dict[str, Any]
457
- image_payload: dict[str, Any] | None = None
458
-
459
- if isinstance(observation, dict):
460
- image_payload = _encode_image_to_base64(observation.get("observation_image"))
461
- sanitized = dict(observation)
462
- sanitized.pop("observation_image", None)
463
- obs_dict = _convert_numpy_to_python(sanitized) or {}
464
- else:
465
- obs_dict = _convert_numpy_to_python(observation) or {}
466
-
467
- if not isinstance(obs_dict, dict):
468
- obs_dict = {"value": obs_dict}
469
-
470
- if image_payload:
471
- obs_dict["observation_image_base64"] = image_payload["data"]
472
- obs_dict["observation_image_format"] = image_payload["format"]
473
- obs_dict["observation_image_width"] = image_payload["width"]
474
- obs_dict["observation_image_height"] = image_payload["height"]
475
- obs_dict["observation_image_data_url"] = image_payload["data_url"]
476
-
477
- return obs_dict
478
-
479
- # ────────────────────────────────────────────────────────────────────
480
- # ReproducibleEnvironment plumbing
481
- # ────────────────────────────────────────────────────────────────────
482
-
483
- async def _serialize_engine(self) -> CrafterEngineSnapshot:
484
- return await self.engine._serialize_engine()
485
-
486
- @classmethod
487
- async def _deserialize_engine(
488
- cls, snapshot: CrafterEngineSnapshot, task_instance: "CrafterTaskInstance"
489
- ) -> "CrafterClassicEnvironment":
490
- eng = await CrafterEngine._deserialize_engine(snapshot, task_instance)
491
- env = cls(task_instance)
492
- env.engine = eng
493
- # CRITICAL: Update the interact tool to use the new engine!
494
- env._interact_tool.engine = eng
495
- return env