auto-coder 0.1.250__py3-none-any.whl → 0.1.252__py3-none-any.whl

{auto_coder-0.1.250.dist-info → auto_coder-0.1.252.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: auto-coder
-Version: 0.1.250
+Version: 0.1.252
 Summary: AutoCoder: AutoCoder
 Author: allwefantasy
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
@@ -26,7 +26,7 @@ Requires-Dist: tabulate
 Requires-Dist: jupyter-client
 Requires-Dist: prompt-toolkit
 Requires-Dist: tokenizers
-Requires-Dist: byzerllm[saas] >=0.1.161
+Requires-Dist: byzerllm[saas] >=0.1.163
 Requires-Dist: patch
 Requires-Dist: diff-match-patch
 Requires-Dist: GitPython

{auto_coder-0.1.250.dist-info → auto_coder-0.1.252.dist-info}/RECORD

@@ -1,17 +1,17 @@
 autocoder/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autocoder/auto_coder.py,sha256=iysR5Bd2IREUKkGkfcmuoLlqn76L70iNKcG-Y-BQvY0,62918
+autocoder/auto_coder.py,sha256=0Bjv8WtImwcObAgBeJ1FZsL6M_wZU04aKMIlLBnCXS8,64568
 autocoder/auto_coder_lang.py,sha256=Rtupq6N3_HT7JRhDKdgCBcwRaiAnyCOR_Gsp4jUomrI,3229
-autocoder/auto_coder_rag.py,sha256=illKgzP2bv-Tq50ujsofJnOHdI4pzr0ALtfR8NHHWdQ,22351
+autocoder/auto_coder_rag.py,sha256=DDAmqw36CO6phtdQuN8LYIbIR3YGdoZw5-pG0LjVxMc,29063
 autocoder/auto_coder_rag_client_mcp.py,sha256=WV7j5JUiQge0x4-B7Hp5-pSAFXLbvLpzQMcCovbauIM,6276
 autocoder/auto_coder_rag_mcp.py,sha256=-RrjNwFaS2e5v8XDIrKR-zlUNUE8UBaeOtojffBrvJo,8521
 autocoder/auto_coder_server.py,sha256=XU9b4SBH7zjPPXaTWWHV4_zJm-XYa6njuLQaplYJH_c,20290
 autocoder/benchmark.py,sha256=Ypomkdzd1T3GE6dRICY3Hj547dZ6_inqJbBJIp5QMco,4423
-autocoder/chat_auto_coder.py,sha256=H83C0FW5q51-Iq0MBAO-9xbr2tCfbF-LVRZddFhmOhM,104869
-autocoder/chat_auto_coder_lang.py,sha256=ffLIVhOrODHamiy4csir-1TD7q7P9NeuZoZ9VaHeuic,15880
+autocoder/chat_auto_coder.py,sha256=oKVoZr66oHNj43SUlqan0i1qBFIleOTrwFDy9iJRgaU,107441
+autocoder/chat_auto_coder_lang.py,sha256=OiwjnOqbiWwbaVuOhehEqdUFPOJbhfvQr3sIq3b0xj4,17455
 autocoder/command_args.py,sha256=9aYJ-AmPxP1sQh6ciw04FWHjSn31f2W9afXFwo8wgx4,30441
 autocoder/lang.py,sha256=U6AjVV8Rs1uLyjFCZ8sT6WWuNUxMBqkXXIOs4S120uk,14511
-autocoder/models.py,sha256=_9Kc8oS_tnnqKzRGnybgOfn1NOey7OXZ8y9qhBmgiB4,5517
-autocoder/version.py,sha256=wNwxrmLzoMunbZEL7IU7ZAvpehOc0XEYZBzjCnolPhU,23
+autocoder/models.py,sha256=uo4mFWb-kdtd-8e2HPfCuXjhkDsE37n1dY6Nt1SU6kk,7625
+autocoder/version.py,sha256=1gGmGBolGnZ8y1dxxyDRAeCnjQZMw6R1feRdZma-dZY,23
 autocoder/agent/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/agent/auto_demand_organizer.py,sha256=NWSAEsEk94vT3lGjfo25kKLMwYdPcpy9e-i21txPasQ,6942
 autocoder/agent/auto_filegroup.py,sha256=CW7bqp0FW1GIEMnl-blyAc2UGT7O9Mom0q66ITz1ckM,6635
@@ -25,30 +25,31 @@ autocoder/agent/project_reader.py,sha256=tWLaPoLw1gI6kO_NzivQj28KbobU2ceOLuppHMb
 autocoder/chat/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/common/JupyterClient.py,sha256=O-wi6pXeAEYhAY24kDa0BINrLYvKS6rKyWe98pDClS0,2816
 autocoder/common/ShellClient.py,sha256=fM1q8t_XMSbLBl2zkCNC2J9xuyKN3eXzGm6hHhqL2WY,2286
-autocoder/common/__init__.py,sha256=xkPcT6OixAN7JLOXcvvXn23Oon0FOmzNXoysxzj6p3k,11890
+autocoder/common/__init__.py,sha256=PjFsST8SG14PiZH4VmMSLGTUz1FqPfcx7GlUmkhT9IQ,11928
 autocoder/common/anything2images.py,sha256=0ILBbWzY02M-CiWB-vzuomb_J1hVdxRcenAfIrAXq9M,25283
 autocoder/common/anything2img.py,sha256=4TREa-sOA-iargieUy7MpyCYVUE-9Mmq0wJtwomPqnE,7662
 autocoder/common/audio.py,sha256=Kn9nWKQddWnUrAz0a_ZUgjcu4VUU_IcZBigT7n3N3qc,7439
-autocoder/common/auto_coder_lang.py,sha256=ljREsdH6VuAeV9cPHPFsQWGxDKIVW8-oFLb8yC4lMQ0,15525
+autocoder/common/auto_coder_lang.py,sha256=9qaluOOt0X6soURfQs27_jw2tIA95yoLdAcACt-Zsf4,16359
 autocoder/common/buildin_tokenizer.py,sha256=L7d5t39ZFvUd6EoMPXUhYK1toD0FHlRH1jtjKRGokWU,1236
 autocoder/common/chunk_validation.py,sha256=BrR_ZWavW8IANuueEE7hS8NFAwEvm8TX34WnPx_1hs8,3030
 autocoder/common/cleaner.py,sha256=NU72i8C6o9m0vXExab7nao5bstBUsfJFcj11cXa9l4U,1089
 autocoder/common/code_auto_execute.py,sha256=4KXGmiGObr_B1d6tzV9dwS6MifCSc3Gm4j2d6ildBXQ,6867
-autocoder/common/code_auto_generate.py,sha256=6rrtdEz8JhlKNutEqlWbgKUlACk90Lcm0UebtuAKEQ0,10320
-autocoder/common/code_auto_generate_diff.py,sha256=o5yeqpc3WXSRWlcLzhlwJNosKo7dcj0CeIsFh6Aibus,16248
-autocoder/common/code_auto_generate_editblock.py,sha256=QdUHUkGaervvQNCY8T2vQ_tfnQX_2kxxu4qq_QW_Nn8,17828
-autocoder/common/code_auto_generate_strict_diff.py,sha256=uteWDEHfIbrnVgwKgqC7qwrIeW0enJCXcHzZGa48yY8,14774
+autocoder/common/code_auto_generate.py,sha256=N_kENWW4eVPKmihpIQ4Q2ivYkugAuCWIPLDCKiarRWg,10818
+autocoder/common/code_auto_generate_diff.py,sha256=QAm-BVJgGl_wbw0VjY2RZigVCwOBO-l1W-90PrYDdOQ,16741
+autocoder/common/code_auto_generate_editblock.py,sha256=FuVH-sP4_Ys2MJk9SkWgHHLAKp51bKSxTwFzd7voilg,18356
+autocoder/common/code_auto_generate_strict_diff.py,sha256=kieVUx2KuR0BGtQHE0Ll7i8e_3Lb4xfFXHIXDttQACE,15274
 autocoder/common/code_auto_merge.py,sha256=8dtnz61l0B5gNbQmx26TZ4_jD825dsnnWtAFD_zs6es,7335
 autocoder/common/code_auto_merge_diff.py,sha256=yocfe8s3Pz6hTGDUl9wRIewY3NcTize_gEla64lsGT0,15331
 autocoder/common/code_auto_merge_editblock.py,sha256=sp7C0fZJMVcNgI8uWy43CKDk7gGXFTkMB9kbP2VdY8k,17485
 autocoder/common/code_auto_merge_strict_diff.py,sha256=9rm0NJ_n6M3LohEX7xl1Jym0xmm8UEYqj_ZTSO3oSlM,9519
-autocoder/common/code_modification_ranker.py,sha256=TycYdtxfNfde2N9-9jEB-9IL2Q3PoHS-k-r9JqZYD6s,6545
+autocoder/common/code_modification_ranker.py,sha256=qfadP9P-iiidCG2A_MjAf3Ca8cMz7YlnN08D_kH6uFc,6447
 autocoder/common/command_completer.py,sha256=SSeb8MDH0JPvfdyW-S2uaHnui4VBDfSQvQPLbv3ORPA,9314
 autocoder/common/command_generator.py,sha256=-hmbD_AnCa5HxL4BznuEfYAf_l8AxU5fAG5F0sM_fuE,2116
 autocoder/common/command_templates.py,sha256=mnB3n8i0yjH1mqzyClEg8Wpr9VbZV44kxky66Zu6OJY,8557
 autocoder/common/const.py,sha256=eTjhjh4Aj4CUzviJ81jaf3Y5cwqsLATySn2wJxaS6RQ,2911
 autocoder/common/files.py,sha256=CguxG9digkWBJpRaILErZmL_G5ryPRahPmPFWGB7X18,1973
 autocoder/common/git_utils.py,sha256=zxgQt2PukabV_21podylAUzTY7Xk60bsQ7MQYw4s-Tg,23234
+autocoder/common/global_cancel.py,sha256=hT7J7J5ChThIhk2x11_v4v9ASIn4HtwyPD26t2s-fwc,418
 autocoder/common/image_to_page.py,sha256=O0cNO_vHHUP-fP4GXiVojShmNqkPnZXeIyiY1MRLpKg,13936
 autocoder/common/interpreter.py,sha256=62-dIakOunYB4yjmX8SHC0Gdy2h8NtxdgbpdqRZJ5vk,2833
 autocoder/common/llm_rerank.py,sha256=FbvtCzaR661Mt2wn0qsuiEL1Y3puD6jeIJS4zg_e7Bs,3260
@@ -56,7 +57,7 @@ autocoder/common/mcp_hub.py,sha256=2ZyJv3Aiv4Y97UHut49oYhIFcu7ICR-mptDEBSgT3uE,1
 autocoder/common/mcp_server.py,sha256=QCFa-15kx7rbNsinwdGFFX2y47pww0fVdI-ldKFSSWI,12267
 autocoder/common/mcp_tools.py,sha256=KsLvRrB6pvmebqd-lDaSH6IBJR0AIxWRE-dtCEG_w9k,12485
 autocoder/common/memory_manager.py,sha256=2ZjYG7BPyvbYalZBF6AM_G5e10Qkw_zrqtD4Zd7GSsQ,3663
-autocoder/common/printer.py,sha256=TIqgGOq5YdWqH4_776QHwHmwfVpubZ9zzUq8rstNxuM,1911
+autocoder/common/printer.py,sha256=P1WU0QjlfnjqTP5uA55GkHZCpFzRPFkc34DMMandreg,2023
 autocoder/common/recall_validation.py,sha256=Avt9Q9dX3kG6Pf2zsdlOHmsjd-OeSj7U1PFBDp_Cve0,1700
 autocoder/common/screenshots.py,sha256=_gA-z1HxGjPShBrtgkdideq58MG6rqFB2qMUJKjrycs,3769
 autocoder/common/search.py,sha256=245iPFgWhMldoUK3CqCP89ltaxZiNPK73evoG6Fp1h8,16518
@@ -73,27 +74,27 @@ autocoder/db/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/db/store.py,sha256=tFT66bP2ZKIqZip-uhLkHRSLaaOAUUDZfozJwcqix3c,1908
 autocoder/dispacher/__init__.py,sha256=YoA64dIxnx4jcE1pwSfg81sjkQtjDkhddkfac1-cMWo,1230
 autocoder/dispacher/actions/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autocoder/dispacher/actions/action.py,sha256=2i9xPgbw6Jn0g3TFIa0JW6FJ21HoOV13J_dYNi9nCEc,20732
+autocoder/dispacher/actions/action.py,sha256=AmckPmTHFKRPTWi4lcw0kAnFw8dyQ9KKoN40JGMuP2s,21971
 autocoder/dispacher/actions/copilot.py,sha256=iMh4ckj9hO5Q-iemF3CStXd7DatWai7Eci5zOlKxK9c,13072
 autocoder/dispacher/actions/plugins/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autocoder/dispacher/actions/plugins/action_regex_project.py,sha256=ht_HWzZt84IEogoFMggnXI6aFFerrsuksVflAkcodfU,5545
+autocoder/dispacher/actions/plugins/action_regex_project.py,sha256=ckTbisMlvwMNHQbrt5WB7pBvf2XAhYQYGH8uyYrvGXU,6060
 autocoder/dispacher/actions/plugins/action_translate.py,sha256=nVAtRSQpdGNmZxg1R_9zXG3AuTv3CHf2v7ODgj8u65c,7727
 autocoder/index/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/index/entry.py,sha256=hXSOi-jPgRBKQB55eqWkT95vxvWrbVHDuIMtDkqjNNw,12232
 autocoder/index/for_command.py,sha256=BFvljE4t6VaMBGboZAuhUCzVK0EitCy_n5D_7FEnihw,3204
-autocoder/index/index.py,sha256=xwh22nY0TtEJMJwYjOUd6xdRZozYXBt47YSwheZSP-4,20679
+autocoder/index/index.py,sha256=hVAIyF10N9hxKMWHA_ibYygGRZYJQZfZxRRrhQhrTvk,21225
 autocoder/index/symbols_utils.py,sha256=CjcjUVajmJZB75Ty3a7kMv1BZphrm-tIBAdOJv6uo-0,2037
 autocoder/index/types.py,sha256=a2s_KV5FJlq7jqA2ELSo9E1sjuLwDB-JJYMhSpzBAhU,596
 autocoder/index/filter/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/index/filter/normal_filter.py,sha256=APu34iSvWhtlLtWgkj8N3Vo4oW1TegtZQq2bwDX_cs4,8031
-autocoder/index/filter/quick_filter.py,sha256=rkRmDOxT2ZF2f2d-xOPtpzVxkSYiw2KWW7g7FY40A0E,9522
+autocoder/index/filter/quick_filter.py,sha256=8R54qtcAAAu0C6TQgJ4WkqkryGOke1mJXrv82vnvMAA,9342
 autocoder/pyproject/__init__.py,sha256=dQ2_7YZ7guybT9BhfxSGn43eLQJGQN2zgeKa6--JlaQ,14403
 autocoder/rag/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/rag/api_server.py,sha256=dRbhAZVRAOlZ64Cnxf4_rKb4iJwHnrWS9Zr67IVORw0,7288
-autocoder/rag/doc_filter.py,sha256=B99Qcy3tcNLuSz2kWbpgfBj2_Igme91zWKOJ2Niq2UY,6652
+autocoder/rag/doc_filter.py,sha256=P2L9u1zqL_igFE6DY5E2r0BY43TrpORjYb_bq7eKT50,6514
 autocoder/rag/document_retriever.py,sha256=5oThtxukGuRFF96o3pHKsk306a8diXbhgSrbqyU2BvM,8894
 autocoder/rag/llm_wrapper.py,sha256=sbDxCANiZyWb_ocqNgqu2oy3c2t8orPNRGleEs-Uwl8,2649
-autocoder/rag/long_context_rag.py,sha256=F-ulTwSZ9ogIGL9jTXDUnI44zStH02EjrztZgRkW9-g,25246
+autocoder/rag/long_context_rag.py,sha256=36ibIdFoOOasYbR1C3WmrfvQs3JB0oYSsk-O4YhTh58,25354
 autocoder/rag/rag_config.py,sha256=8LwFcTd8OJWWwi1_WY4IzjqgtT6RyE2j4PjxS5cCTDE,802
 autocoder/rag/rag_entry.py,sha256=6TKtErZ0Us9XSV6HgRKXA6yR3SiZGPHpynOKSaR1wgE,2463
 autocoder/rag/raw_rag.py,sha256=yS2Ur6kG0IRjhCj2_VonwxjY_xls_E62jO5Gz5j2nqE,2952
@@ -102,7 +103,7 @@ autocoder/rag/simple_directory_reader.py,sha256=LkKreCkNdEOoL4fNhc3_hDoyyWTQUte4
 autocoder/rag/simple_rag.py,sha256=I902EUqOK1WM0Y2WFd7RzDJYofElvTZNLVCBtX5A9rc,14885
 autocoder/rag/token_checker.py,sha256=jc76x6KWmvVxds6W8juZfQGaoErudc2HenG3sNQfSLs,2819
 autocoder/rag/token_counter.py,sha256=C-Lwc4oIjJpZDEqp9WLHGOe6hb4yhrdJpMtkrtp_1qc,2125
-autocoder/rag/token_limiter.py,sha256=dtIxCtHswZ2ut-XKbx8_SiWyv-xqnR1WAIcmh6f8Ktw,11137
+autocoder/rag/token_limiter.py,sha256=5JznJ-h1bG5x2Plx5H0MGkaAHo2GiD7CDIJmeZWWk_8,11015
 autocoder/rag/types.py,sha256=WPgLpUTwbk0BAikyDOc0NOEwV5k73myF38zWdOuYdC4,2499
 autocoder/rag/utils.py,sha256=x5L8gskxUUGkVEP5K-0C-iYntE84asuDXOCqdzdNUoA,4956
 autocoder/rag/variable_holder.py,sha256=PFvBjFcR7-fNDD4Vcsc8CpH2Te057vcpwJMxtrfUgKI,75
@@ -128,7 +129,7 @@ autocoder/utils/_markitdown.py,sha256=RU88qn4eZfYIy0GDrPxlI8oYXIypbi63VRJjdlnE0V
 autocoder/utils/coder.py,sha256=rK8e0svQBe0NOP26dIGToUXgha_hUDgxlWoC_p_r7oc,5698
 autocoder/utils/conversation_store.py,sha256=sz-hhY7sttPAUOAQU6Pze-5zJc3j0_Emj22dM_0l5ro,1161
 autocoder/utils/llm_client_interceptors.py,sha256=FEHNXoFZlCjAHQcjPRyX8FOMjo6rPXpO2AJ2zn2KTTo,901
-autocoder/utils/llms.py,sha256=YH2hJIkHUEBOz93nXzJmWUIHC9oXFlHDjE8DF3NP2q4,2252
+autocoder/utils/llms.py,sha256=RBOSzH6xhedISzmiQMGH_swmVfY-QSTe5Cm1ZZGBhNE,2948
 autocoder/utils/log_capture.py,sha256=I-bsJFLWoGUiX-GKoZsH9kWJCKSV7ZlUnRt7jh-fOL0,1548
 autocoder/utils/multi_turn.py,sha256=unK9OpqVRbK6uIcTKXgggX2wNmyj7s5eyEAQ2xUwHoM,88
 autocoder/utils/operate_config_api.py,sha256=99YAKsuUFLPwrRvj0CJal_bAPgyiXWMma6ZKMU56thw,5790
@@ -138,13 +139,14 @@ autocoder/utils/request_event_queue.py,sha256=r3lo5qGsB1dIjzVQ05dnr0z_9Z3zOkBdP1
 autocoder/utils/request_queue.py,sha256=nwp6PMtgTCiuwJI24p8OLNZjUiprC-TsefQrhMI-yPE,3889
 autocoder/utils/rest.py,sha256=hLBhr78y-WVnV0oQf9Rxc22EwqF78KINkScvYa1MuYA,6435
 autocoder/utils/tests.py,sha256=BqphrwyycGAvs-5mhH8pKtMZdObwhFtJ5MC_ZAOiLq8,1340
+autocoder/utils/thread_utils.py,sha256=uXc9HEr1kSTi491lc-BhFUsbc0wb3ikltZNJkL7Ocz4,8195
 autocoder/utils/types.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/utils/auto_coder_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autocoder/utils/auto_coder_utils/chat_stream_out.py,sha256=5kKOrPy1GWZ6qHjni-YNJJb2x0rD7_3Idl2M9sWxVfU,10481
+autocoder/utils/auto_coder_utils/chat_stream_out.py,sha256=wdunUiqCe6LntrEdxqx2iClwNGXhyL5kB52Y_4qHyNo,11014
 autocoder/utils/chat_auto_coder_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-auto_coder-0.1.250.dist-info/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
-auto_coder-0.1.250.dist-info/METADATA,sha256=ZAZ-ZHjFf6PUo8PYcj62mgVAEO2CXzJBRSA1z-bppc8,2616
-auto_coder-0.1.250.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
-auto_coder-0.1.250.dist-info/entry_points.txt,sha256=0nzHtHH4pNcM7xq4EBA2toS28Qelrvcbrr59GqD_0Ak,350
-auto_coder-0.1.250.dist-info/top_level.txt,sha256=Jqc0_uJSw2GwoFQAa9iJxYns-2mWla-9ok_Y3Gcznjk,10
-auto_coder-0.1.250.dist-info/RECORD,,
+auto_coder-0.1.252.dist-info/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
+auto_coder-0.1.252.dist-info/METADATA,sha256=I9oxYdimJoBFwT8BdU5eU6PUhuesvVq-CTZpF5s55GU,2616
+auto_coder-0.1.252.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
+auto_coder-0.1.252.dist-info/entry_points.txt,sha256=0nzHtHH4pNcM7xq4EBA2toS28Qelrvcbrr59GqD_0Ak,350
+auto_coder-0.1.252.dist-info/top_level.txt,sha256=Jqc0_uJSw2GwoFQAa9iJxYns-2mWla-9ok_Y3Gcznjk,10
+auto_coder-0.1.252.dist-info/RECORD,,

autocoder/auto_coder.py

@@ -6,6 +6,7 @@ from autocoder.common import git_utils, code_auto_execute
 from autocoder.utils.llm_client_interceptors import token_counter_interceptor
 from autocoder.db.store import Store
 
+from autocoder.utils.llms import get_llm_names
 from autocoder.utils.queue_communicate import (
     queue_communicate,
     CommunicateEvent,
@@ -45,6 +46,7 @@ from autocoder import models as models_module
 from autocoder.common.utils_code_auto_generate import stream_chat_with_continue
 from autocoder.utils.auto_coder_utils.chat_stream_out import stream_out
 from autocoder.common.printer import Printer
+from autocoder.rag.token_counter import count_tokens
 console = Console()
 
 
@@ -764,6 +766,23 @@ def main(input_args: Optional[List[str]] = None):
                 )
                 llm.setup_sub_client("planner_model", planner_model)
 
+            if args.commit_model:
+                model_name = args.commit_model.strip()
+                model_info = models_module.get_model_by_name(model_name)
+                commit_model = byzerllm.SimpleByzerLLM(default_model_name=model_name)
+                commit_model.deploy(
+                    model_path="",
+                    pretrained_model_type=model_info["model_type"],
+                    udf_name=model_name,
+                    infer_params={
+                        "saas.base_url": model_info["base_url"],
+                        "saas.api_key": model_info["api_key"],
+                        "saas.model": model_info["model_name"],
+                        "saas.is_reasoning": model_info["is_reasoning"]
+                    }
+                )
+                llm.setup_sub_client("commit_model", commit_model)    
+
             if args.designer_model:
                 model_name = args.designer_model.strip()
                 model_info = models_module.get_model_by_name(model_name)
@@ -874,6 +893,11 @@ def main(input_args: Optional[List[str]] = None):
                 designer_model.setup_default_model_name(args.designer_model)
                 llm.setup_sub_client("designer_model", designer_model)
 
+            if args.commit_model:
+                commit_model = byzerllm.ByzerLLM()
+                commit_model.setup_default_model_name(args.commit_model)
+                llm.setup_sub_client("commit_model", commit_model)
+
     else:
         llm = None
 
@@ -1333,6 +1357,13 @@ def main(input_args: Optional[List[str]] = None):
                 reviewer = AutoReviewCommit(llm=chat_llm, args=args)
                 v = reviewer.review_commit(query=args.query,conversations=loaded_conversations)
             else:                
+                # 预估token数量
+                estimated_input_tokens = count_tokens(json.dumps(loaded_conversations, ensure_ascii=False))
+                printer = Printer()
+                printer.print_in_terminal("estimated_chat_input_tokens", style="yellow",
+                                  estimated_input_tokens=estimated_input_tokens
+                                  )
+
                 v = stream_chat_with_continue(
                     llm=chat_llm,
                     conversations=loaded_conversations,
@@ -1340,9 +1371,7 @@ def main(input_args: Optional[List[str]] = None):
                 )              
 
 
-            model_name = getattr(chat_llm, 'default_model_name', None)
-            if not model_name:
-                model_name = "unknown(without default model name)"    
+            model_name = ",".join(get_llm_names(chat_llm))
             
             assistant_response, last_meta = stream_out(
                     v, 
@@ -1356,10 +1385,13 @@ def main(input_args: Optional[List[str]] = None):
             if last_meta:
                 elapsed_time = time.time() - start_time
                 printer = Printer()
+                speed = last_meta.generated_tokens_count / elapsed_time
                 printer.print_in_terminal("stream_out_stats", 
                                     elapsed_time=elapsed_time,
+                                    first_token_time=last_meta.first_token_time,
                                     input_tokens=last_meta.input_tokens_count,
-                                    output_tokens=last_meta.generated_tokens_count)
+                                    output_tokens=last_meta.generated_tokens_count,
+                                    speed=round(speed, 2))
             
             chat_history["ask_conversation"].append(
                 {"role": "assistant", "content": assistant_response}

autocoder/auto_coder_rag.py

@@ -32,7 +32,10 @@ if platform.system() == "Windows":
     init()
 
 
-def initialize_system():
+def initialize_system(args):
+    if args.product_mode == "lite":
+        return
+
     print(f"\n\033[1;34m{get_message('initializing')}\033[0m")
 
     def print_status(message, status):
@@ -316,13 +319,30 @@ def main(input_args: Optional[List[str]] = None):
         action="store_true",
         help="Whether to return responses without contexts. only works when pro plugin is installed",
     )
-    serve_parser.add_argument(
-        "--data_cells_max_num",
+
+    serve_parser.add_argument("--data_cells_max_num",
         type=int,
         default=2000,
         help="Maximum number of data cells to process",
     )
 
+    serve_parser.add_argument(
+        "--product_mode",
+        type=str,
+        default="pro",
+        help="The mode of the auto-coder.rag, lite/pro default is pro",
+    )
+    serve_parser.add_argument(
+        "--lite",
+        action="store_true",
+        help="Run in lite mode (equivalent to --product_mode=lite)",
+    )
+    serve_parser.add_argument(
+        "--pro",
+        action="store_true",
+        help="Run in pro mode (equivalent to --product_mode=pro)",
+    )
+
     serve_parser.add_argument(
         "--recall_model",
         default="",
@@ -373,6 +393,22 @@ def main(input_args: Optional[List[str]] = None):
     # Tools command
     tools_parser = subparsers.add_parser("tools", help="Various tools")
     tools_subparsers = tools_parser.add_subparsers(dest="tool", help="Available tools")
+    tools_parser.add_argument(
+        "--product_mode",
+        type=str,
+        default="pro",
+        help="The mode of the auto-coder.rag, lite/pro default is pro",
+    )
+    tools_parser.add_argument(
+        "--lite",
+        action="store_true",
+        help="Run in lite mode (equivalent to --product_mode=lite)",
+    )
+    tools_parser.add_argument(
+        "--pro",
+        action="store_true",
+        help="Run in pro mode (equivalent to --product_mode=pro)",
+    )
 
     # Count tool
     count_parser = tools_subparsers.add_parser("count", help="Count tokens in a file")
@@ -431,8 +467,15 @@ def main(input_args: Optional[List[str]] = None):
             benchmark_byzerllm(args.model, args.parallel, args.rounds, args.query)
 
     elif args.command == "serve":
+         # Handle lite/pro flags
+        if args.lite:
+            args.product_mode = "lite"
+        elif args.pro:
+            args.product_mode = "pro"
+
         if not args.quick:
-            initialize_system()
+            initialize_system(args)
+       
         server_args = ServerArgs(
             **{
                 arg: getattr(args, arg)
@@ -448,7 +491,11 @@ def main(input_args: Optional[List[str]] = None):
             }
         )
 
-        if auto_coder_args.enable_hybrid_index:
+
+        if auto_coder_args.enable_hybrid_index and args.product_mode == "lite":
+            raise Exception("Hybrid index is not supported in lite mode")
+
+        if auto_coder_args.enable_hybrid_index and args.product_mode == "pro":
             # 尝试连接storage
             try:
                 from byzerllm.apps.byzer_storage.simple_api import ByzerStorage
@@ -460,36 +507,107 @@ def main(input_args: Optional[List[str]] = None):
                     "When enable_hybrid_index is true, ByzerStorage must be started"
                 )
                 logger.error("Please run 'byzerllm storage start' first")
-                return
-        else:
-            byzerllm.connect_cluster(address=args.ray_address)
-        llm = byzerllm.ByzerLLM()
-        llm.setup_default_model_name(args.model)
-
-        # Setup sub models if specified
-        if args.recall_model:
-            recall_model = byzerllm.ByzerLLM()
-            recall_model.setup_default_model_name(args.recall_model)
-            llm.setup_sub_client("recall_model", recall_model)
-
-        if args.chunk_model:
-            chunk_model = byzerllm.ByzerLLM()
-            chunk_model.setup_default_model_name(args.chunk_model)
-            llm.setup_sub_client("chunk_model", chunk_model)
+                return                        
+        
+        
 
-        if args.qa_model:
-            qa_model = byzerllm.ByzerLLM()
-            qa_model.setup_default_model_name(args.qa_model)
-            llm.setup_sub_client("qa_model", qa_model)
+        if args.product_mode == "pro":
+            byzerllm.connect_cluster(address=args.ray_address)
+            llm = byzerllm.ByzerLLM()
+            llm.skip_nontext_check = True
+            llm.setup_default_model_name(args.model)
+
+            # Setup sub models if specified
+            if args.recall_model:
+                recall_model = byzerllm.ByzerLLM()
+                recall_model.setup_default_model_name(args.recall_model)
+                recall_model.skip_nontext_check = True
+                llm.setup_sub_client("recall_model", recall_model)
+
+            if args.chunk_model:
+                chunk_model = byzerllm.ByzerLLM()
+                chunk_model.setup_default_model_name(args.chunk_model)
+                llm.setup_sub_client("chunk_model", chunk_model)
+
+            if args.qa_model:
+                qa_model = byzerllm.ByzerLLM()
+                qa_model.setup_default_model_name(args.qa_model)
+                qa_model.skip_nontext_check = True
+                llm.setup_sub_client("qa_model", qa_model)
+
+            # 当启用hybrid_index时,检查必要的组件
+            if auto_coder_args.enable_hybrid_index:
+                if not llm.is_model_exist("emb"):
+                    logger.error(
+                        "When enable_hybrid_index is true, an 'emb' model must be deployed"
+                    )
+                    return
+                llm.setup_default_emb_model_name("emb")
+
+        elif args.product_mode == "lite":
+            from autocoder import models as models_module
+            model_info = models_module.get_model_by_name(args.model)
+            llm = byzerllm.SimpleByzerLLM(default_model_name=args.model)
+            llm.deploy(
+                model_path="",
+                pretrained_model_type=model_info["model_type"],
+                udf_name=args.model,
+                infer_params={
+                    "saas.base_url": model_info["base_url"],
+                    "saas.api_key": model_info["api_key"],
+                    "saas.model": model_info["model_name"],
+                    "saas.is_reasoning": model_info["is_reasoning"]
+                }
+            )
 
-        # 当启用hybrid_index时,检查必要的组件
-        if auto_coder_args.enable_hybrid_index:
-            if not llm.is_model_exist("emb"):
-                logger.error(
-                    "When enable_hybrid_index is true, an 'emb' model must be deployed"
+            # Setup sub models if specified
+            if args.recall_model:
+                model_info = models_module.get_model_by_name(args.recall_model)
+                recall_model = byzerllm.SimpleByzerLLM(default_model_name=args.recall_model)
+                recall_model.deploy(
+                    model_path="",
+                    pretrained_model_type=model_info["model_type"],
+                    udf_name=args.recall_model,
+                    infer_params={
+                        "saas.base_url": model_info["base_url"],
+                        "saas.api_key": model_info["api_key"],
+                        "saas.model": model_info["model_name"],
+                        "saas.is_reasoning": model_info["is_reasoning"]
+                    }
                 )
-                return
-            llm.setup_default_emb_model_name("emb")
+                llm.setup_sub_client("recall_model", recall_model)
+
+            if args.chunk_model:
+                model_info = models_module.get_model_by_name(args.chunk_model)
+                chunk_model = byzerllm.SimpleByzerLLM(default_model_name=args.chunk_model)
+                chunk_model.deploy(
+                    model_path="",
+                    pretrained_model_type=model_info["model_type"],
+                    udf_name=args.chunk_model,
+                    infer_params={
+                        "saas.base_url": model_info["base_url"],
+                        "saas.api_key": model_info["api_key"],
+                        "saas.model": model_info["model_name"],
+                        "saas.is_reasoning": model_info["is_reasoning"]
+                    }
+                )
+                llm.setup_sub_client("chunk_model", chunk_model)
+
+            if args.qa_model:
+                model_info = models_module.get_model_by_name(args.qa_model)
+                qa_model = byzerllm.SimpleByzerLLM(default_model_name=args.qa_model)
+                qa_model.deploy(
+                    model_path="",
+                    pretrained_model_type=model_info["model_type"],
+                    udf_name=args.qa_model,
+                    infer_params={
+                        "saas.base_url": model_info["base_url"],
+                        "saas.api_key": model_info["api_key"],
+                        "saas.model": model_info["model_name"],
+                        "saas.is_reasoning": model_info["is_reasoning"]
+                    }
+                )
+                llm.setup_sub_client("qa_model", qa_model)
 
         if server_args.doc_dir:
             auto_coder_args.rag_type = "simple"
@@ -577,16 +695,61 @@ def main(input_args: Optional[List[str]] = None):
             count_tokens(args.tokenizer_path, args.file)
         elif args.tool == "recall":
             from .common.recall_validation import validate_recall
-
-            llm = byzerllm.ByzerLLM.from_default_model(args.model)
+            from autocoder import models as models_module
+
+            # Handle lite/pro flags
+            if args.lite:
+                args.product_mode = "lite"
+            elif args.pro:
+                args.product_mode = "pro"
+
+            if args.product_mode == "pro":
+                llm = byzerllm.ByzerLLM.from_default_model(args.model)
+            else:  # lite mode
+                model_info = models_module.get_model_by_name(args.model)
+                llm = byzerllm.SimpleByzerLLM(default_model_name=args.model)
+                llm.deploy(
+                    model_path="",
+                    pretrained_model_type=model_info["model_type"],
+                    udf_name=args.model,
+                    infer_params={
+                        "saas.base_url": model_info["base_url"],
+                        "saas.api_key": model_info["api_key"],
+                        "saas.model": model_info["model_name"],
+                        "saas.is_reasoning": model_info["is_reasoning"]
+                    }
+                )
 
             content = None if not args.content else [args.content]
             result = validate_recall(llm, content=content, query=args.query)
             print(f"Recall Validation Result:\n{result}")
+
         elif args.tool == "chunk":
             from .common.chunk_validation import validate_chunk
+            from autocoder import models as models_module
+            
+            if args.lite:
+                args.product_mode = "lite"
+            elif args.pro:
+                args.product_mode = "pro"
+
+            if args.product_mode == "pro":
+                llm = byzerllm.ByzerLLM.from_default_model(args.model)
+            else:  # lite mode
+                model_info = models_module.get_model_by_name(args.model)
+                llm = byzerllm.SimpleByzerLLM(default_model_name=args.model)
+                llm.deploy(
+                    model_path="",
+                    pretrained_model_type=model_info["model_type"],
+                    udf_name=args.model,
+                    infer_params={
+                        "saas.base_url": model_info["base_url"],
+                        "saas.api_key": model_info["api_key"],
+                        "saas.model": model_info["model_name"],
+                        "saas.is_reasoning": model_info["is_reasoning"]
+                    }
+                )
 
-            llm = byzerllm.ByzerLLM.from_default_model(args.model)
             content = None if not args.content else [args.content]
             result = validate_chunk(llm, content=content, query=args.query)
             print(f"Chunk Model Validation Result:\n{result}")