llm.rb 12.0.0 → 12.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e042f679a5f682feed6726430dd1b28fd6728c9223cbd0e0499c92e8952a06e6
-  data.tar.gz: 0cb3f55e70a321bea3f1903261142bf3aa842fface494331a548f684aac520d8
+  metadata.gz: 20d8c7c9c3d9d270f371781b28e7537a5c6921cff138cfc895bdbf0042003d07
+  data.tar.gz: 11b1abd397f1b13afadf4fe56e9670037a7c6ee586e945fd6a909505a4114122
 SHA512:
-  metadata.gz: ba64ba9b87d1343f9d94a06b97cb6cc0ec0ca5d7a8c4b235e52c9ed489c5670848b44e0549e838b01b12be89d19005ba571a6dd9557a8f2f5f4e62a3ec018ec9
-  data.tar.gz: 81a489afa510edf608b74a3a83abc16b6d81c6f59f614df6460f638dbcf1e0a539d8d2b6b2e5a98a3772d5c9d7c7233d7b231bbf447dda7cfcf8df1861819c8d
+  metadata.gz: d99ad697a3b3d5d13a0804ab3cea569ffbf5dc88d853aabe7461c4ae59020519e140aa0de1c2b26336b03998212410449da238ea8df5f6a190012f026093464c
+  data.tar.gz: dd6314a5e1e7ba3899d1ab13ba4f076508481bc2e24d3502bfa5cc4ebfb66a2203772ffc9acb4987b741eea413997493b8f85a2a7a841cf768e80d9b412886a8

data/CHANGELOG.md

@@ -13,7 +13,199 @@
 > Changelog <br>
 > a [r.uby.dev](https://r.uby.dev) project
 
-## Unreleased
+## What's next
+
+Changes since `v12.2.0`.
+
+## v12.2.0
+
+Changes since `v12.1.0`.
+
+This release adds Mistral as a new provider with chat completions, streaming, tool calls,
+structured outputs, file/image attachments, and embeddings support. It introduces
+the `trace:` option to `LLM::Agent#repl` for keeping the tracer active during
+interactive sessions.
+
+Several fixes land for the Google provider (generationConfig parameter leakage),
+`LLM::Context#tracer=` (always assigning nil), `LLM::Provider#with_tracer(nil)`
+(nil fallback), and `LLM::Context#repair!` (dropping Struct returns).
+
+The default HTTP timeout has been increased from 60s to 180s to better accommodate
+reasoning models and large structured outputs, and the Anthropic default model has
+been updated to `claude-opus-4-8`. Model metadata has been refreshed across
+Anthropic, AWS Bedrock, DeepInfra, Google, and xAI, with Mistral model data added
+to the registry.
+
+### Add
+
+* **Add `trace:` option to `LLM::Agent#repl`** <br>
+  `LLM::Agent#repl` now accepts a `trace:` keyword argument. By default
+  the tracer is disabled for the duration of the repl session to prevent
+  curses UI interference from output written to `$stdout` or `$stderr`.
+  Set `trace: true` to keep the tracer active during the session, which
+  is useful when the tracer writes to a file rather than the terminal.
+
+* **Add a new provider: LLM::Mistral** <br>
+  [Mistral](https://mistral.ai) is now supported through its
+  OpenAI-compatible API. The provider supports chat completions,
+  streaming, tool calls, structured output (schema), file/image
+  attachments, and embeddings. Use `LLM.mistral(...)` to create a
+  provider instance.
+
+* **Add `LLM.mistral(...)` convenience method** <br>
+  A new top-level accessor (`LLM.mistral`) returns an `LLM::Mistral`
+  provider instance, matching the pattern used by other providers.
+
+### Fix
+
+* **Fix Google `generationConfig` parameter leakage** <br>
+  Fix a bug in the Google provider where non-generation parameters
+  (`role`, `model`, `messages`, `stream`) were leaking into the
+  `generationConfig` object alongside legitimate generation config
+  parameters such as `temperature`. Non-config parameters are now
+  filtered out before constructing `generationConfig`.
+
+* **Fix `LLM::Context#tracer=` always assigning `nil`** <br>
+  The `LLM::Context#tracer=` setter had a bug where it always assigned
+  `nil` regardless of the tracer value passed. It now correctly assigns
+  the given tracer or falls back to `LLM::Tracer::Null`.
+
+* **Fix `LLM::Provider#with_tracer(nil)` fallback** <br>
+  `LLM::Provider#with_tracer(nil)` now falls back to
+  `LLM::Tracer::Null` instead of setting a `nil` tracer directly.
+
+* **Fix `LLM::Context#repair!` dropping `Struct` returns** <br>
+  `LLM::Context#repair!` used `[*prompt]` to wrap the prompt before
+  grepping for return objects. Since `LLM::Function::Return` is a
+  `Struct`, the splat operator expanded it into its member values
+  instead of wrapping it, causing the grep to silently drop returns.
+  The fix wraps both sources in an array before flattening.
+
+### Change
+
+* **Increase default provider timeout from 60s to 180s** <br>
+  The default HTTP timeout for all providers has been increased from
+  60 to 180 seconds to better accommodate long-running requests such
+  as reasoning models and large structured outputs.
+
+* **Change Anthropic default model to `claude-opus-4-8`** <br>
+  The default Anthropic chat model has been updated from
+  `claude-sonnet-4-20250514` to `claude-opus-4-8`, reflecting the
+  latest model release from Anthropic.
+
+* **Refresh model metadata** <br>
+  Update model listings, pricing, and capabilities for Anthropic,
+  AWS Bedrock, DeepInfra, Google, and xAI. Add Mistral model data
+  to the registry.
+
+## v12.1.0
+
+Changes since `v12.0.0`.
+
+This release adds `LLM::Agent#repl` with a curses-based
+interactive read-eval-print loop. It requires the optional
+dependency `curses` and it is probably the most notable
+feature in this release.
+
+Multiple _opt-in_ tools have been added to the `llm/tools/*.rb`
+directory. They serve as examples and as general-purpose tools
+that happen to power the repository's agents.
+
+The BSL license has been extended to grant additional free waivers
+for non-profits, charities and for companies with 50 or less
+employees.
+
+Other changes include small-ish bug fixes. <br>
+As always, see the changelog details for a thorough overview.
+
+### Add
+
+* **Add `LLM::Agent#repl`** <br>
+  Add a curses-based read-eval-print loop for `LLM::Agent` that lets
+  developers interact with an agent after it has been set up or has
+  performed a task. It is similar to `binding.pry`: once you exit,
+  you can continue with the rest of your program. It requires the
+  `curses` gem.
+
+* **Add `#tracer=` setter on Provider, Context and Agent** <br>
+  `LLM::Provider`, `LLM::Context` and `LLM::Agent` can now configure
+  the tracer after initialization via the `#tracer=` setter. It accepts
+  a subclass of `LLM::Tracer` or `nil` to disable the tracer.
+
+* **Add `LLM::Tool::Shell`** <br>
+  Add a built-in shell tool that can run a command with arguments. <br>
+  It must be required explicitly with `require "llm/tools/shell"` and
+  requires the `test-cmd.rb` gem.
+
+* **Add `LLM::Tool::ReadFile`** <br>
+  Add a built-in tool for reading the contents of a file, with optional
+  `start` and `stop` line offsets. <br>
+  It must be required explicitly with `require "llm/tools/read_file"`.
+
+* **Add `LLM::Tool::Chdir`** <br>
+  Add a built-in tool for changing the current working directory. <br>
+  It must be required explicitly with `require "llm/tools/chdir"`.
+
+* **Add `LLM::Tool::Git`** <br>
+  Add a built-in tool that can perform git actions (`log`, `diff`,
+  `show`, `commit`, `checkout`, `branch`). <br>
+  It must be required explicitly with `require "llm/tools/git"` and requires the `test-cmd.rb` gem.
+
+* **Add `LLM::Tool::Rg`** <br>
+  Add a built-in tool that wraps the `rg` (ripgrep) command for
+  recursively searching the current directory for patterns. <br>
+  It must be required explicitly with `require "llm/tools/rg"` and requires the `test-cmd.rb` gem.
+
+* **Add `LLM::Tool::SwapText`** <br>
+  Add a built-in tool that can replace an exact snippet of text in a
+  file with a new piece of text. <br>
+  It must be required explicitly with `require "llm/tools/swap_text"`.
+
+* **Add `LLM::Tool::Pwd`** <br>
+  Add a built-in tool that returns the current working directory. <br>
+  It must be required explicitly with `require "llm/tools/pwd"`.
+
+* **Add `LLM::Tool::WriteFile`** <br>
+  Add a built-in tool that can write a given string to a given file
+  path. <br>
+  It must be required explicitly with `require "llm/tools/write_file"`.
+
+* **Add `LLM::Tool::Mkdir`** <br>
+  Add a built-in tool that can create a tree of new directories. <br>
+  It must be required explicitly with `require "llm/tools/mkdir"` and
+  requires the `test-cmd.rb` gem.
+
+### Change
+
+* **Extend BSL additional use grant** <br>
+  The Business Source License additional use grant has been extended to
+  include non-profits, charities, and companies with 50 or fewer
+  employees, in addition to the existing personal, education, and
+  evaluation uses.
+
+* **Change LlamaCpp default port (8080 => 8013)** <br>
+  The default port for the LlamaCpp provider has changed from `8080` to
+  `8013` since llamacpp itself defaults to that port.
+
+* **Change LlamaCpp default model to `nil`** <br>
+  The default model for LlamaCpp is now `nil`, letting whatever model
+  is served by the llamacpp server act as the default. Previously it
+  defaulted to `qwen3`.
+
+### Fix
+
+* **Fix `LLM::Agent.tools` Symbol resolution** <br>
+  When an agent defined tools via `tools :method_name`, the resolved
+  symbol was incorrectly forwarded as `[:method_name]` (an array) to
+  `LLM::Context`. This fix copies the same pattern used by other
+  attribute resolvers (e.g., `skills`) so a single Symbol is resolved
+  through the agent instance correctly.
+
+* **Encode strings as UTF-8 in the JSON adapter** <br>
+  The `json` gem will reject BINARY-encoded strings from version 3 and
+  beyond. The `LLM::JSONAdapter.dump` method now walks serialized data
+  and encodes every string into UTF-8, using `String#scrub` to replace
+  bytes that are not valid UTF-8.
 
 ## v12.0.0
 

data/LICENSE

@@ -10,10 +10,11 @@ Licensor: Robert Gleeson
 Licensed Work: llm.rb
 
 Additional Use Grant:
-Free for personal use.
-Free for education.
-Free for evaluation, development, and testing.
-Commercial production use requires a commercial license.
+- Free for personal use
+- Free for students and their teachers
+- Free for evaluation, development, and testing
+- Free for non-profits and charities
+- Free for companies with less than or equal to 50 employees
 
 Change Date:
 Four years after the first public release of each specific version of the

data/README.md

@@ -10,7 +10,7 @@
   </a>
 </p>
 
-> A [r.uby.dev](https://r.uby.dev) project.
+> A [r.uby.dev](https://r.uby.dev/llm) project.
 
 Welcome to the canonical llm.rb repository.
 
@@ -23,7 +23,7 @@ that are opt-in.
 ## Features
 
 The runtime supports OpenAI, OpenAI-compatible endpoints, Anthropic, Google
-Gemini, DeepSeek, DeepInfra, xAI, Z.ai, AWS Bedrock, Ollama, and llama.cpp.
+Gemini, Mistral, DeepSeek, DeepInfra, xAI, Z.ai, AWS Bedrock, Ollama, and llama.cpp.
 It has first-class support for streaming, tool calls,  MCP
 and A2A, embeddings, vector stores and the RAG pattern.
 
@@ -117,6 +117,24 @@ agent = LLM::Agent.new(llm, stream: MyStream.new)
 agent.talk "Explain Ruby fibers."
 ```
 
+#### LLM::REPL
+
+The [LLM::Agent#repl](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html#repl-instance_method)
+method allows an agent to spawn a read-eval-print loop
+that can be useful during the development and operation
+of agents. It can be used to debug tool calls, confirm an
+agent has done what was expected, or improve the agent
+by asking questions about what it has done up to that point.
+
+This feature requires that the [curses](https://github.com/ruby/curses)
+library is installed and available to require.
+
+```ruby
+llm = LLM.deepseek(key: ENV["KEY"])
+agent = LLM::Agent.new(llm)
+agent.repl
+```
+
 #### LLM::MCP
 
 The Model Context Protocol (MCP) has first-class support
@@ -247,28 +265,37 @@ agent.talk "perform research"
 <summary>What providers does llm.rb support?</summary>
 <br>
 <p>
-China-based
 
-* DeepSeek
-* zAI
+**Cloud**
+
+The following cloud-based providers are available to choose from. <br>
+In no particular order:
+
+🇺🇸 OpenAI <br>
+🇺🇸 DeepInfra <br>
+🇺🇸 xAI <br>
+🇺🇸 Google (Gemini) <br>
+🇺🇸 AWS bedrock <br>
+🇺🇸 Anthropic <br>
+🇨🇳 DeepSeek <br>
+🇨🇳 zAI <br>
+🇪🇺 Mistral <br>
 
-US-based
+**Weights**
 
-* OpenAI
-* Google (Gemini)
-* xAI
-* AWS bedrock
-* DeepInfra
-* Anthropic
+The following providers provide access to open-weight models. <br>
+In no particular order:
 
-Openweights
+🇺🇸 DeepInfra <br>
+🇺🇸 AWS bedrock <br>
+🇨🇳 DeepSeek <br>
+🇨🇳 zAI <br>
+🇪🇺 Mistral <br>
 
-* DeepSeek
-* zAI
-* DeepInfra
-* AWS bedrock
+**Local**
 
-Host your own
+The following providers can be run locally on your own hardware. <br>
+In no particular order:
 
 * Ollama
 * Llamacpp
@@ -335,3 +362,14 @@ Each version converts to the [BSD Zero Clause](https://choosealicense.com/licens
 four years after its first public release.
 <br>
 Contact [robert@r.uby.dev](mailto:robert@r.uby.dev) for a commercial license.
+
+### Waivers
+
+Waivers are automatically granted for: <br>
+
+  * Personal use
+  * Students
+  * Teachers
+  * Evaluation, development, and testing
+  * Non-profits and charities
+  * Companies with less than or equal to 50 employees