Release 0.1.0 🎉

0.1.0 release

Built packages

• pypi: archgw/0.1.0
• pypi: archgw_modelserver/0.1.1
• docker image: 64ec02e9d1d5 (katanemo/archgw:latest)

What's Changed

• Add initial scaffolding by @junr03 in #1
• add checks on pr to ensure code compiles and passes static analysis checks. by @aayushwhiz in #4
• Add stats API by @junr03 in #5
• Parse katanemo config using serde/yaml package by @adilhafeez in #6
• Setup pre-commit so it runs locally before every git push by 🎉@adilhafeez in #12
• add embedding store by @adilhafeez in #10
• Add testing to GitHub workflow by @aayushwhiz in #15
• Add initial logic to send prompts to LLM API by @junr03 in #9
• Delete example code, and move prod code into functional separation by @junr03 in #16
• Address comments from clean up PR by @junr03 in #17
• Use open-message-format to serialize and deserialize embeddings api by @adilhafeez in #18
• update pre-commit to only run fmt on intelligent-prompt-gateway package by @adilhafeez in #22
• refactor code a bit by @adilhafeez in #21
• Add initial integration style tests by @junr03 in #20
• Improve error handling by @junr03 in #23
• Add basic workflow logic by @adilhafeez in #24
• Clean up imports by @junr03 in #25
• Remove unnecessary clones by @junr03 in #26
• Remove top level container in config and use snake-case for config by @adilhafeez in #27
• update open-message-format@1e838f3 by @adilhafeez in #30
• Improve demo experience by @adilhafeez in #28
• Simplify Entity struct by @junr03 in #33
• Add support for local llm (mistral 7b) by @adilhafeez in #31
• Add Ratelimit config by @junr03 in #32
• Add the Ratelimit module by @junr03 in #34
• Move shared types into their own crate by @junr03 in #41
• Add Ratelimit on request tokens by @junr03 in #44
• Fix wasm load error by @junr03 in #45
• Add function calling support using bolt-fc-1b by @adilhafeez in #35
• More idiomatic conversion by @junr03 in #47
• Add small clarification to function calling demo by @junr03 in #48
• Resetting instead of resuming by @junr03 in #39
• Improve prompt target intent matching by @adilhafeez in #51
• rename envoy-1 => bolt-1 by @adilhafeez in #53
• Improve service names by @adilhafeez in #54
• add bolt guard config by @adilhafeez in #55
• Propagate enums from prompt-target to bolt-fc by @adilhafeez in #56
• Add ability to stream a response by @junr03 in #50
• add ability to override default values from config by @adilhafeez in #58
• send all tools when sending request to arch-fc by @adilhafeez in #59
• Add ability to define clusters in config by @adilhafeez in #61
• demos for network copilot and sql analyzer by @Sampreeth-sarma in #57
• use openai standard response in arch-fc and in gradio client by @adilhafeez in #62
• fix webui url and dependencies by @adilhafeez in #66
• Delete unused envoy.yaml file by @junr03 in #65
• remove embeddings config from config by @adilhafeez in #64
• Include param default in parameters by @adilhafeez in #68
• fix similarity bug by @Sampreeth-sarma in #63
• move demo functions out of model_server by @adilhafeez in #67
• Docs branch - v1 of our tech docs by @salmanap in #69
• [Kan-103] add support toxic/jailbreak model by @cotran2 in #49
• Improve response handling by @adilhafeez in #71
• add details on how to use grafana dashboards by @adilhafeez in #72
• improve cluster not configured error by @adilhafeez in #73
• remove guard config json by @cotran2 in #70
• Adil/fix salman docs by @salmanap in #75
• fix demos code by @adilhafeez in #76
• Remove OMF by @adilhafeez in #78
• Add support for multiple LLM Providers by @junr03 in #60
• improve logging of api failure by @adilhafeez in #79
• add readme and docker build by @adilhafeez in #81
• send history to bolt fc model by @adilhafeez in #84
• Added Float type to the function parameter values by @Sampreeth-sarma in #77
• Fix bug in PromptGuard configuration by @junr03 in #80
• Revert "Revert "Add support for multiple LLM Providers (#60)"" by @junr03 in #83
• Integrate Arch-Function-Calling-1.5B model by @adilhafeez in #85
• V1 docs push by @salmanap in #86
• expose access logs from envoy by @adilhafeez in #89
• add bolt support by @adilhafeez in #90
• Add supported parameter type, validation and tests by @adilhafeez in #88
• Salmanap/docs v1 push by @salmanap in #92
• rename envoyfilter => arch by @adilhafeez in #91
• add support for 3b model by @adilhafeez in #96
• changes prometheus target to arch so that data collection works by @aayushwhiz in #98
• add precommit check by @adilhafeez in #97
• update config by @adilhafeez in #93
• Remove unnecessary envoy.yaml by @junr03 in #99
• remove method type by @adilhafeez in #101
• Rename bolt_config to arch_config by @junr03 in #100
• Fold function_resolver into model_server by @adilhafeez in #103
• add messages in params when making api calls by @adilhafeez in #104
• arch schema validator by @adilhafeez in #105
• simplify developer getting started experience by @salmanap in #102
• rename archgw_model_sever => model_server by @adilhafeez in #106
• ArchFC endpoint integration by @cotran2 in #94
• update path for arch_config.yaml file by @adilhafeez in #107
• change root to config in config_generator by @aayushwhiz in #108
• fix fc integration by @cotran2 in #110
• change default stat_prefix from ingress_http to arch by @aayushwhiz in #109
• add support for default target by @adilhafeez in #111
• Add the ability to use LLM Providers from the Arch config by @junr03 in #112
• Remove optional PromptGuards from Stream Context by @junr03 in #113
• archgw cli by @salmanap in #117
• Cotran/hallucination integration by @cotran2 in #115
• Remove hardcoded Access Key Envvars by @junr03 in #118
• load_models checks for device before getting the BGE or NLI model loa… by @salmanap in #119
• fix embeddings not found bug by @adilhafeez in #120
• [BUG FIX] Add missing chat completions request hint by @junr03 in #116
• Clean up Embeddings Store by @junr03 in #121
• fix pkg_resources error by @adilhafeez in #123
• Salmanap/fix config generator by @salmanap in #124
• don't compute embeddings for names and other fixes see description by @adilhafeez in #126
• Build arch docs by @salmanap in #128
• Doc Update by @nehcgs in #129
• added index.html and made minor README.md edits by @salmanap in #130
• model server build by @salmanap in #127
• add files for local development by @adilhafeez in #133
• Serialize tool calls for Arch FC by @adilhafeez in #131
• fixing docs so that GH pages picks up the right CNAME for DNS by @salmanap in #135
• formating and mointoring change by @cotran2 in #136
• Update arch_config.yaml for function calling demo by @nehcgs in #137
• Implement Client trait for StreamContext by @junr03 in #134
• add system prompt by @adilhafeez in #138
• system prompt (keep system prompt at the top) by @adilhafeez in #139
• Hallucination integration with rust by @cotran2 in #122
• Salmanap/fix demos by @salmanap in #140
• ensure that we can call the new api.fc.archgw.com url, logging fixes … by @salmanap in #142
• Fix errors and improve Doc by @nehcgs in #143
• Cotran/onnx conversion by @cotran2 in #145
• Cotran/prompt guard doc by @cotran2 in #147
• Split listener by @adilhafeez in #141
• fix dependcy + log info by @cotran2 in #148
• ensure that tracing is optional in arch_config by @adilhafeez in #149
• add days and units to api server by @adilhafeez in #150
• Fix model_server poetry lock file by @adilhafeez in #151
• Salmanap/fix network agent demo by @salmanap in #153
• updated all demo READMes and minor doc changes by @salmanap in #154
• minor fixes to README by @salmanap in #156
• lint + formating with black by @cotran2 in #158
• Fix gpu dependency and only leverage onnx when GPU is available by @cotran2 in #157
• llm listener split by @adilhafeez in #155
• fixed cli to use poetry as well. this way we make it easy to have the… by @salmanap in #160
• update cli and update docs by @adilhafeez in #161
• dont send default target to archfc by @adilhafeez in #166
• Update model_server by @nehcgs in #164
• change nli model by @cotran2 in #167
• make ratelimit section optional by @adilhafeez in #168
• updated archgw cli to pull from archgw_modelserver from pypi by @salmanap in #169
• update access logs docs by @adilhafeez in #170
• fix for linux by @cotran2 in #175
• fix prometheus target and update dashboard to be more full. by @aayushwhiz in #165
• Improve cli by @adilhafeez in #179
• Update doc by @nehcgs in #178
• minor changes to the README to highlight what Arch is designed to do … by @salmanap in #182
• ensuring that logo appears in dark mode by @salmanap in #184
• Salmanap/fix discord invite by @salmanap in #185
• concatenate history of user messages for hallucination by @cotran2 in #177
• split wasm filter by @adilhafeez in #186
• Refactor model server hardware config + add unit tests to load/request to the server by @cotran2 in #189
• Split arch wasm filter code into prompt and llm gateway filters by @adilhafeez in #190
• update lock file by @adilhafeez in #192
• fix cli models and logs by @salmanap in #196
• Add cargo workspace to allow rust-analyzer to work correctly by @junr03 in #197
• Code refactor and some improvements - see description by @adilhafeez in #194
• Update vscode workspce by @adilhafeez in #199
• Send back developer error correctly by @adilhafeez in #195
• Update arch Dockerfile by @junr03 in #200
• Pass tool call and app function response back in metadata by @adilhafeez in #193
• update observability and dashboards by @adilhafeez in #198
• Add separate util for hallucination and add tests for it by @adilhafeez in #203
• refactor prompt gateway by @adilhafeez in #204
• Improve logging by @adilhafeez in #209
• Cotran/hallucination by @cotran2 in #208
• HR agent demo by @salmanap in #206
• fixing discord link and moving contributing guide to root by @salmanap in #215
• docs: update README.md by @eltociear in #220
• Fix to issue #171 rename arch_messages => messages by @itsazibfarooq in #211
• Update chatbot UI and update hallucination check by @nehcgs in #218
• fixed typos in arch_config.yaml file based on issue #221 by @salmanap in #223
• add README for hr_agent by @salmanap in #224
• fix model server stop process by @cotran2 in #217
• fixed typos in intro to arch docs by @salmanap in #225
• Add support for streaming and fixes few issues (see description) by @junr03 in #202
• Improve Gradio UI and fix arch_state bug by @adilhafeez in #227
• moving chatbot-ui in demos and out of root project structure by @salmanap in #228
• Updated hr_agent to be full stack: gradio + fastAPI by @salmanap in #235
• Use large github action machine to run e2e tests by @adilhafeez in #230
• Salmanap/fix all demos by @salmanap in #238
• release 0.1.0 by @adilhafeez in #239

New Contributors

• @aayushwhiz made their first contribution in #4
• @Sampreeth-sarma made their first contribution in #57
• @salmanap made their first contribution in #69
• @nehcgs made their first contribution in #129
• @eltociear made their first contribution in #220
• @itsazibfarooq made their first contribution in #211

Full Changelog: https://github.com/katanemo/arch/commits/0.1.0