Commit Graph

43 Commits

Author SHA1 Message Date
Sam 7ca21cc329
FEATURE: first class support for OpenRouter (#1011)
* FEATURE: first class support for OpenRouter

This new implementation supports picking quantization and provider pref

Also:

- Improve logging for summary generation
- Improve error message when contacting LLMs fails

* Better support for full screen artifacts on iPad

Support back button to close full screen
2024-12-10 05:59:19 +11:00
Sam d56ed53eb1
FIX: cancel functionality regressed (#938)
The cancel messaging was not floating correctly to the HTTP call leading to impossible to cancel completions 

This is now fully tested as well.
2024-11-21 17:51:45 +11:00
Sam 755b63f31f
FEATURE: Add support for Mistral models (#919)
Adds support for mistral models (pixtral and mistral large now have presets)

Also corrects token accounting in AWS bedrock models
2024-11-19 17:28:09 +11:00
Sam 0d7f353284
FEATURE: AI artifacts (#898)
This is a significant PR that introduces AI Artifacts functionality to the discourse-ai plugin along with several other improvements. Here are the key changes:

1. AI Artifacts System:
   - Adds a new `AiArtifact` model and database migration
   - Allows creation of web artifacts with HTML, CSS, and JavaScript content
   - Introduces security settings (`strict`, `lax`, `disabled`) for controlling artifact execution
   - Implements artifact rendering in iframes with sandbox protection
   - New `CreateArtifact` tool for AI to generate interactive content

2. Tool System Improvements:
   - Adds support for partial tool calls, allowing incremental updates during generation
   - Better handling of tool call states and progress tracking
   - Improved XML tool processing with CDATA support
   - Fixes for tool parameter handling and duplicate invocations

3. LLM Provider Updates:
   - Updates for Anthropic Claude models with correct token limits
   - Adds support for native/XML tool modes in Gemini integration
   - Adds new model configurations including Llama 3.1 models
   - Improvements to streaming response handling

4. UI Enhancements:
   - New artifact viewer component with expand/collapse functionality
   - Security controls for artifact execution (click-to-run in strict mode)
   - Improved dialog and response handling
   - Better error management for tool execution

5. Security Improvements:
   - Sandbox controls for artifact execution
   - Public/private artifact sharing controls
   - Security settings to control artifact behavior
   - CSP and frame-options handling for artifacts

6. Technical Improvements:
   - Better post streaming implementation
   - Improved error handling in completions
   - Better memory management for partial tool calls
   - Enhanced testing coverage

7. Configuration:
   - New site settings for artifact security
   - Extended LLM model configurations
   - Additional tool configuration options

This PR significantly enhances the plugin's capabilities for generating and displaying interactive content while maintaining security and providing flexible configuration options for administrators.
2024-11-19 09:22:39 +11:00
Sam 823e8ef490
FEATURE: partial tool call support for OpenAI and Anthropic (#908)
Implement streaming tool call implementation for Anthropic and Open AI.

When calling:

llm.generate(..., partial_tool_calls: true) do ...
Partials may contain ToolCall instances with partial: true, These tool calls are partially populated with json partially parsed.

So for example when performing a search you may get:

ToolCall(..., {search: "hello" })
ToolCall(..., {search: "hello world" })

The library used to parse json is:

https://github.com/dgraham/json-stream

We use a fork cause we need access to the internal buffer.

This prepares internals to perform partial tool calls, but does not implement it yet.
2024-11-14 06:58:24 +11:00
Sam e817b7dc11
FEATURE: improve tool support (#904)
This re-implements tool support in DiscourseAi::Completions::Llm #generate

Previously tool support was always returned via XML and it would be the responsibility of the caller to parse XML

New implementation has the endpoints return ToolCall objects.

Additionally this simplifies the Llm endpoint interface and gives it more clarity. Llms must implement

decode, decode_chunk (for streaming)

It is the implementers responsibility to figure out how to decode chunks, base no longer implements. To make this easy we ship a flexible json decoder which is easy to wire up.

Also (new)

    Better debugging for PMs, we now have a next / previous button to see all the Llm messages associated with a PM
    Token accounting is fixed for vllm (we were not correctly counting tokens)
2024-11-12 08:14:30 +11:00
Sam 059d3b6fd2
FEATURE: better logging for automation reports (#853)
A new feature_context json column was added to ai_api_audit_logs

This allows us to store rich json like context on any LLM request
made.

This new field now stores automation id and name.

Additionally allows llm_triage to specify maximum number of tokens

This means that you can limit the cost of llm triage by scanning only
first N tokens of a post.
2024-10-23 16:49:56 +11:00
Sam 5b9add0ac8
FEATURE: add a SambaNova LLM provider (#797)
Note, at the moment the context window is quite small, it is
mainly useful as a helper backend or hyde generator
2024-09-12 11:28:08 +10:00
Roman Rizzi bed044448c
DEV: Remove old code now that features rely on LlmModels. (#729)
* DEV: Remove old code now that features rely on LlmModels.

* Hide old settings and migrate persona llm overrides

* Remove shadowing special URL + seeding code. Use srv:// prefix instead.
2024-07-30 13:44:57 -03:00
Sam 8b81ff45b8
FIX: switch off native tools on Anthropic Claude Opus (#659)
Native tools do not work well on Opus.

Chain of Thought prompting means it consumes enormous amounts of
tokens and has poor latency.

This commit introduce and XML stripper to remove various chain of
thought XML islands from anthropic prompts when tools are involved.

This mean Opus native tools is now functions (albeit slowly)

From local testing XML just works better now.

Also fixes enum support in Anthropic native tools
2024-06-07 10:52:01 -03:00
Sam 3993c685e1
FEATURE: anthropic function calling (#654)
Adds support for native tool calling (both streaming and non streaming) for Anthropic.

This improves general tool support on the Anthropic models.
2024-06-06 08:34:23 +10:00
Sam baf88e7cfc
FEATURE: improve logging by including llm name (#640)
Log the language model name when logging api requests
2024-05-27 16:46:01 +10:00
Roman Rizzi 3a9080dd14
FEATURE: Test LLM configuration (#634) 2024-05-21 13:35:50 -03:00
Roman Rizzi 1d786fbaaf
FEATURE: Set endpoint credentials directly from LlmModel. (#625)
* FEATURE: Set endpoint credentials directly from LlmModel.

Drop Llama2Tokenizer since we no longer use it.

* Allow http for custom LLMs

---------

Co-authored-by: Rafael Silva <xfalcox@gmail.com>
2024-05-16 09:50:22 -03:00
Sam 8eee6893d6
FEATURE: GPT4o support and better auditing (#618)
- Introduce new support for GPT4o (automation / bot / summary / helper)
- Properly account for token counts on OpenAI models
- Track feature that was used when generating AI completions
- Remove custom llm support for summarization as we need better interfaces to control registration and de-registration
2024-05-14 13:28:46 +10:00
Roman Rizzi e22194f321
HACK: Llama3 support for summarization/AI helper. (#616)
There are still some limitations to which models we can support with the `LlmModel` class. This will enable support for Llama3 while we sort those out.
2024-05-13 15:54:42 -03:00
Sam 514823daca
FIX: streaming broken in bedrock when chunks are not aligned (#609)
Also

- Stop caching llm list - this cause llm list in persona to be incorrect
- Add more UI to debug screen so you can properly see raw response
2024-05-09 12:11:50 +10:00
Roman Rizzi 4f1a3effe0
REFACTOR: Migrate Vllm/TGI-served models to the OpenAI format. (#588)
Both endpoints provide OpenAI-compatible servers. The only difference is that Vllm doesn't support passing tools as a separate parameter. Even if the tool param is supported, it ultimately relies on the model's ability to handle native functions, which is not the case with the models we have today.

As a part of this change, we are dropping support for StableBeluga/Llama2 models. They don't have a chat_template, meaning the new API can translate them.

These changes let us remove some of our existing dialects and are a first step in our plan to support any LLM by defining them as data-driven concepts.

 I rewrote the "translate" method to use a template method and extracted the tool support strategies into its classes to simplify the code.

Finally, these changes bring support for Ollama when running in dev mode. It only works with Mistral for now, but it will change soon..
2024-05-07 10:02:16 -03:00
Sam a223d18f1a
FIX: more robust function call support (#581)
For quite a few weeks now, some times, when running function calls
on Anthropic we would get a "stray" - "calls" line.

This has been enormously frustrating!

I have been unable to find the source of the bug so instead decoupled
the implementation and create a very clear "function call normalizer"

This new class is extensively tested and guards against the type of
edge cases we saw pre-normalizer.

This also simplifies the implementation of "endpoint" which no longer
needs to handle all this complex logic.
2024-04-19 06:54:54 +10:00
Sam 7f16d3ad43
FEATURE: Cohere Command R support (#558)
- Added Cohere Command models (Command, Command Light, Command R, Command R Plus) to the available model list
- Added a new site setting `ai_cohere_api_key` for configuring the Cohere API key
- Implemented a new `DiscourseAi::Completions::Endpoints::Cohere` class to handle interactions with the Cohere API, including:
   - Translating request parameters to the Cohere API format
   - Parsing Cohere API responses 
   - Supporting streaming and non-streaming completions
   - Supporting "tools" which allow the model to call back to discourse to lookup additional information
- Implemented a new `DiscourseAi::Completions::Dialects::Command` class to translate between the generic Discourse AI prompt format and the Cohere Command format
- Added specs covering the new Cohere endpoint and dialect classes
- Updated `DiscourseAi::AiBot::Bot.guess_model` to map the new Cohere model to the appropriate bot user

In summary, this PR adds support for using the Cohere Command family of models with the Discourse AI plugin. It handles configuring API keys, making requests to the Cohere API, and translating between Discourse's generic prompt format and Cohere's specific format. Thorough test coverage was added for the new functionality.
2024-04-11 07:24:17 +10:00
Sam f62703760f
FEATURE: add Claude 3 sonnet/haiku support for Amazon Bedrock (#534)
This PR consolidates the  implements new Anthropic Messages interface for Bedrock Claude endpoints and adds support for the new Claude 3 models (haiku, opus, sonnet).

Key changes:
- Renamed `AnthropicMessages` and `Anthropic` endpoint classes into a single `Anthropic` class (ditto for ClaudeMessages -> Claude)
- Updated `AwsBedrock` endpoints to use the new `/messages` API format for all Claude models
- Added `claude-3-haiku`, `claude-3-opus` and `claude-3-sonnet` model support in both Anthropic and AWS Bedrock endpoints
- Updated specs for the new consolidated endpoints and Claude 3 model support

This refactor removes support for old non messages API which has been deprecated by anthropic
2024-03-19 06:48:46 +11:00
Sam 2ad743d246
FEATURE: Add GitHub Helper AI Bot persona and tools (#513)
Introduces a new AI Bot persona called 'GitHub Helper' which is specialized in assisting with GitHub-related tasks and questions. It includes the following key changes:

- Implements the GitHub Helper persona class with its system prompt and available tools
   
- Adds three new AI Bot tools for GitHub interactions:
  - github_file_content: Retrieves content of files from a GitHub repository
  - github_pull_request_diff: Retrieves the diff for a GitHub pull request
  - github_search_code: Searches for code in a GitHub repository
    
- Updates the AI Bot dialects to support the new GitHub tools

- Implements multiple function calls for standard tool dialect
2024-03-08 06:37:23 +11:00
Sam 8b382d6098
FEATURE: support for claude opus and sonnet (#508)
This provides new support for messages API from Claude.

It is required for latest model access.

Also corrects implementation of function calls.

* Fix message interleving

* fix broken spec

* add new models to automation
2024-03-06 06:04:37 +11:00
Sam c02794cf2e
FIX: support multiple tool calls (#502)
* FIX: support multiple tool calls

Prior to this change we had a hard limit of 1 tool call per llm
round trip. This meant you could not google multiple things at
once or perform searches across two tools.

Also:

- Hint when Google stops working
- Log topic_id / post_id when performing completions

* Also track id for title
2024-03-02 07:53:21 +11:00
Roman Rizzi 94ba0dadc2
SECURITY: Place a SSRF protection when calling services from the plugin. (#485)
The Faraday adapter and `FinalDestionation::HTTP` will protect us from admin-initiated SSRF attacks when interacting with the external services powering this plugin features.:
2024-02-21 17:14:50 -03:00
Sam cec4251b00
DEV: improve error bedrock error messages (#454)
When bedrock rate limits it returns a 200 BUT also returns a JSON
document with the error.

Previously we had no special case here so we complained about nil

New code properly logs the problem
2024-02-01 08:01:07 -03:00
Roman Rizzi 0634b85a81
UX: Validations to LLM-backed features (except AI Bot) (#436)
* UX: Validations to Llm-backed features (except AI Bot)

This change is part of an ongoing effort to prevent enabling a broken feature due to lack of configuration. We also want to explicit which provider we are going to use. For example, Claude models are available through AWS Bedrock and Anthropic, but the configuration differs.

Validations are:

* You must choose a model before enabling the feature.
* You must turn off the feature before setting the model to blank.
* You must configure each model settings before being able to select it.

* Add provider name to summarization options

* vLLM can technically support same models as HF

* Check we can talk to the selected model

* Check for Bedrock instead of anthropic as a site could have both creds setup
2024-01-29 16:04:25 -03:00
Roman Rizzi 5bdf3dc1f4
DEV: Stop using shared_examples for endpoint specs (#430) 2024-01-17 15:08:49 -03:00
Jarek Radosz 6b8a57d957
DEV: Update linting (#423)
Co-authored-by: Keegan George <kgeorge13@gmail.com>
2024-01-13 00:28:06 +01:00
Sam 8df966e9c5
FEATURE: smooth streaming of AI responses on the client (#413)
This PR introduces 3 things:

1. Fake bot that can be used on local so you can test LLMs, to enable on dev use:

SiteSetting.ai_bot_enabled_chat_bots = "fake"

2. More elegant smooth streaming of progress on LLM completion

This leans on JavaScript to buffer and trickle llm results through. It also amends it so the progress dot is much 
more consistently rendered

3. It fixes the Claude dialect 

Claude needs newlines **exactly** at the right spot, amended so it is happy 

---------

Co-authored-by: Martin Brennan <martin@discourse.org>
2024-01-11 15:56:40 +11:00
Roman Rizzi 971e03bdf2
FEATURE: AI Bot Gemini support. (#402)
It also corrects the syntax around tool support, which was wrong.

Gemini doesn't want us to include messages about previous tool invocations, so I had to shuffle around some code to send the response it generated from those invocations instead. For this, I created the "multi_turn" context, which bundles all the context involved in the interaction.
2024-01-04 18:15:34 -03:00
Roman Rizzi f9d7d7f5f0
DEV: AI bot migration to the Llm pattern. (#343)
* DEV: AI bot migration to the Llm pattern.

We added tool and conversation context support to the Llm service in discourse-ai#366, meaning we met all the conditions to migrate this module.

This PR migrates to the new pattern, meaning adding a new bot now requires minimal effort as long as the service supports it. On top of this, we introduce the concept of a "Playground" to separate the PM-specific bits from the completion, allowing us to use the bot in other contexts like chat in the future. Commands are called tools, and we simplified all the placeholder logic to perform updates in a single place, making the flow more one-wayish.

* Followup fixes based on testing

* Cleanup unused inference code

* FIX: text-based tools could be in the middle of a sentence

* GPT-4-turbo support

* Use new LLM API
2024-01-04 10:44:07 -03:00
Sam 03fc94684b
FIX: AI helper not working correctly with mixtral (#399)
* FIX: AI helper not working correctly with mixtral

This PR introduces a new function on the generic llm called #generate

This will replace the implementation of completion!

#generate introduces a new way to pass temperature, max_tokens and stop_sequences

Then LLM implementers need to implement #normalize_model_params to
ensure the generic names match the LLM specific endpoint

This also adds temperature and stop_sequences to completion_prompts
this allows for much more robust completion prompts

* port everything over to #generate

* Fix translation

- On anthropic this no longer throws random "This is your translation:"
- On mixtral this actually works

* fix markdown table generation as well
2024-01-04 09:53:47 -03:00
Roman Rizzi 4182af230a
FIX: Correctly translate and read tools for Claude and Chat GPT. (#393)
I tested against the live models for the AI bot migration. It ensures Open AI's tool syntax is correct and we can correctly read the replies.
:
2024-01-02 11:21:13 -03:00
Rafael dos Santos Silva 5db7bf6e68
Mixtral (#376)
Add both Mistral and Mixtral support. Also includes vLLM-openAI inference support.

Co-authored-by: Roman Rizzi <rizziromanalejandro@gmail.com>
2023-12-26 14:49:55 -03:00
Sam af2e692761
FIX: under certain conditions we would get duplicate data from llm (#373)
Previously endpoint/base would `+=` decoded_chunk to leftover

This could lead to cases where the leftover buffer had duplicate
previously processed data

Fix ensures we properly skip previously decoded data.
2023-12-20 14:28:05 -03:00
Roman Rizzi e0bf6adb5b
DEV: Tool support for the LLM service. (#366)
This PR adds tool support to available LLMs. We'll buffer tool invocations and return them instead of making users of this service parse the response.

It also adds support for conversation context in the generic prompt. It includes bot messages, user messages, and tool invocations, which we'll trim to make sure it doesn't exceed the prompt limit, then translate them to the correct dialect.

Finally, It adds some buffering when reading chunks to handle cases when streaming is extremely slow.:M
2023-12-18 18:06:01 -03:00
Rafael dos Santos Silva 83744bf192
FEATURE: Support for Gemini in AiHelper / Search / Summarization (#358) 2023-12-15 14:32:01 -03:00
Rafael dos Santos Silva d8267d8da0
FIX: Many fixes for huggingface and llama2 inference (#335) 2023-12-06 11:22:42 -03:00
Sam 6ddc17fd61
DEV: port directory structure to Zeitwerk (#319)
Previous to this change we relied on explicit loading for a files in Discourse AI.

This had a few downsides:

- Busywork whenever you add a file (an extra require relative)
- We were not keeping to conventions internally ... some places were OpenAI others are OpenAi
- Autoloader did not work which lead to lots of full application broken reloads when developing.

This moves all of DiscourseAI into a Zeitwerk compatible structure.

It also leaves some minimal amount of manual loading (automation - which is loading into an existing namespace that may or may not be there)

To avoid needing /lib/discourse_ai/... we mount a namespace thus we are able to keep /lib pointed at ::DiscourseAi

Various files were renamed to get around zeitwerk rules and minimize usage of custom inflections

Though we can get custom inflections to work it is not worth it, will require a Discourse core patch which means we create a hard dependency.
2023-11-29 15:17:46 +11:00
Roman Rizzi f26adf2cf6
FIX: Use XML tags in generate_titles prompt. (#322)
We must ensure we can isolate titles, and the models sometimes ignore the example we give them.

Additionally, anons can generate HyDE posts, so we need to check if user is nil when attempting to log requests.
2023-11-28 12:52:22 -03:00
Roman Rizzi 2e7c5f047d
DEV: Don't attempt to update log if completion request fails. (#321)
We already log the request failure when we raise the exception.
2023-11-28 11:15:12 -03:00
Roman Rizzi 3064d4c288
REFACTOR: Summarization and HyDE now use an LLM abstraction. (#297)
* DEV: One LLM abstraction to rule them all

* REFACTOR: HyDE search uses new LLM abstraction

* REFACTOR: Summarization uses the LLM abstraction

* Updated documentation and made small fixes. Remove Bedrock claude-2 restriction
2023-11-23 12:58:54 -03:00