diff --git a/CHANGELOG.md b/CHANGELOG.md --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -7,6 +7,92 @@ ## [Unreleased] +## [baikai 0.4.1.0] - 2026-07-20 + +### Changed + +- Version bump only; no library API or code changes. Released so the umbrella + release tag `baikai-0.4.1.0` names a fresh core version alongside the breaking + `baikai-claude` / `baikai-openai` 0.4.0.0 releases, matching the tag + convention downstream consumers pin against. + +## [baikai-claude 0.4.0.0] - 2026-07-20 + +### Changed + +- **Breaking:** `claudeCliCommand` now takes the `Options` record and forwards + `Options.thinking` to batch `claude -p` as `--effort ` (`minimal` + collapses to `low`, matching the interactive launcher and the claude CLI's + lack of a `minimal` value). `thinking = Nothing` emits no effort flag, keeping + existing argv byte-for-byte. The added parameter is a PVP-major signature + change. + +## [baikai-openai 0.4.0.0] - 2026-07-20 + +### Changed + +- **Breaking:** `codexCliCommand` now takes the `Options` record and forwards + `Options.thinking` to `codex exec` as `-c model_reasoning_effort=` for + all six effort levels. `thinking = Nothing` emits no override, keeping + existing argv byte-for-byte. The added parameter is a PVP-major signature + change. + +## [baikai 0.4.0.0] - 2026-07-20 + +### Added + +- Added `ThinkingXHigh` and `ThinkingMax` to the exported `ThinkingLevel` + vocabulary and added a defaulted `InteractiveLaunchRequest.effort` field. + Extending the closed sum type is a PVP-major API change for downstream + exhaustive matches. + +## [baikai-claude 0.3.0.2] - 2026-07-20 + +### Added + +- Added `--effort` rendering to interactive Claude Code launches and preserved + `xhigh` / `max` on native adaptive Anthropic API requests, with larger fixed + budgets for manual-thinking models. + +### Changed + +- Bumped the internal `baikai` dependency bound to `^>=0.4.0` for the + baikai 0.4.0.0 release. + +## [baikai-openai 0.3.0.2] - 2026-07-20 + +### Added + +- Added `model_reasoning_effort` overrides to interactive Codex launches and + preserved `xhigh` / `max` in native OpenAI request JSON; non-native + OpenAI-compatible request shapes continue to clamp them to `high`. + +### Changed + +- Bumped the internal `baikai` dependency bound to `^>=0.4.0` for the + baikai 0.4.0.0 release. + +## [baikai-trace-otel 0.3.0.2] - 2026-07-20 + +### Changed + +- Bumped the internal `baikai` dependency bound to `^>=0.4.0` for the + baikai 0.4.0.0 release. No API changes. + +## [baikai-effectful 0.3.0.2] - 2026-07-20 + +### Changed + +- Bumped the internal `baikai` dependency bound to `^>=0.4.0` for the + baikai 0.4.0.0 release. No API changes. + +## [baikai-kit 0.1.0.3] - 2026-07-20 + +### Changed + +- Bumped the internal `baikai` dependency bound to `^>=0.4.0` for the + baikai 0.4.0.0 release. No API changes. + ## [baikai 0.3.1.0] - 2026-07-15 ### Added diff --git a/baikai.cabal b/baikai.cabal --- a/baikai.cabal +++ b/baikai.cabal @@ -1,6 +1,6 @@ cabal-version: 3.4 name: baikai -version: 0.3.1.0 +version: 0.4.1.0 synopsis: Unified Haskell interface for multiple AI providers description: baikai provides a unified, provider-agnostic Haskell interface for working @@ -151,6 +151,7 @@ InteractiveSpec StreamSpec SurfaceSpec + ThinkingLevelSpec TraceSpec UsageSpec diff --git a/gen/GenModelsCore.hs b/gen/GenModelsCore.hs --- a/gen/GenModelsCore.hs +++ b/gen/GenModelsCore.hs @@ -322,6 +322,7 @@ dropTrailingEmpty ( header ++ concatMap (renderEntry . snd) entries + ++ renderAllModels (map (ident . snd) entries) ) where dropTrailingEmpty = \case @@ -371,6 +372,25 @@ "import Data.Ratio ((%))", "" ] + +-- | Render the @allModels@ binding: every generated model in one list, in the +-- same (identifier-sorted) order as the declarations above. Consumers can fold +-- or filter this instead of hand-listing bindings, so a catalog change flows +-- through automatically. +renderAllModels :: [Text] -> [Text] +renderAllModels idents = + "-- | Every model in the generated catalog, in declaration order." + : "allModels :: [Model]" + : "allModels =" + : case idents of + [] -> [" []"] + _ -> + [ " " <> prefix <> ident <> suffix + | (position, ident) <- zip [0 :: Int ..] idents, + let prefix = if position == 0 then "[ " else " ", + let suffix = if position == length idents - 1 then "" else "," + ] + ++ [" ]"] renderEntry :: GeneratedEntry -> [Text] renderEntry g = diff --git a/src/Baikai/Interactive.hs b/src/Baikai/Interactive.hs --- a/src/Baikai/Interactive.hs +++ b/src/Baikai/Interactive.hs @@ -15,6 +15,7 @@ extraDirs, safety, extraArgs, + effort, InteractiveSafety (..), CodexSandboxMode (..), CodexApprovalPolicy (..), @@ -31,6 +32,7 @@ where import Baikai.Prelude +import Baikai.ThinkingLevel (ThinkingLevel) import System.Exit (ExitCode) -- | Local interactive provider families that Baikai knows how to @@ -56,7 +58,8 @@ workingDir :: !(Maybe FilePath), extraDirs :: ![FilePath], safety :: !InteractiveSafety, - extraArgs :: ![Text] + extraArgs :: ![Text], + effort :: !(Maybe ThinkingLevel) } deriving stock (Eq, Show, Generic) @@ -98,7 +101,8 @@ workingDir = Nothing, extraDirs = [], safety = DefaultSafety, - extraArgs = [] + extraArgs = [], + effort = Nothing } interactiveLaunchResult :: InteractiveProvider -> ExitCode -> InteractiveLaunchResult diff --git a/src/Baikai/Models/Generated.hs b/src/Baikai/Models/Generated.hs --- a/src/Baikai/Models/Generated.hs +++ b/src/Baikai/Models/Generated.hs @@ -841,3 +841,43 @@ headers = Map.empty, compat = CompatNone } + +-- | Every model in the generated catalog, in declaration order. +allModels :: [Model] +allModels = + [ anthropic_claude_fable_5, + anthropic_claude_haiku_4_5, + anthropic_claude_opus_4_5, + anthropic_claude_opus_4_6, + anthropic_claude_opus_4_7, + anthropic_claude_opus_4_8, + anthropic_claude_sonnet_4_5, + anthropic_claude_sonnet_4_6, + anthropic_claude_sonnet_5, + deepseek_deepseek_chat, + deepseek_deepseek_reasoner, + openai_gpt_4_1, + openai_gpt_4_1_mini, + openai_gpt_4_1_nano, + openai_gpt_4o, + openai_gpt_4o_mini, + openai_gpt_5, + openai_gpt_5_1, + openai_gpt_5_2, + openai_gpt_5_4, + openai_gpt_5_4_mini, + openai_gpt_5_4_nano, + openai_gpt_5_5, + openai_gpt_5_6, + openai_gpt_5_6_luna, + openai_gpt_5_6_sol, + openai_gpt_5_6_terra, + openai_gpt_5_mini, + openai_gpt_5_nano, + openai_o1, + openai_o3, + openai_o3_mini, + openai_o4_mini, + openrouter_anthropic_claude_sonnet_4, + openrouter_openai_gpt_4o_mini + ] diff --git a/src/Baikai/ThinkingLevel.hs b/src/Baikai/ThinkingLevel.hs --- a/src/Baikai/ThinkingLevel.hs +++ b/src/Baikai/ThinkingLevel.hs @@ -2,10 +2,10 @@ -- | Provider-agnostic reasoning-effort preference. -- --- Each provider maps the value to its own primitive: OpenAI uses --- @reasoning_effort: "minimal" | "low" | "medium" | "high"@; --- Anthropic uses a token budget on its @thinking@ field. --- 'thinkingTokenBudget' provides the recommended budget for the +-- Each provider maps the value to its own primitive. The canonical +-- rendering spans @minimal@ through @max@, while provider boundaries +-- may preserve, translate, or clamp values according to their APIs. +-- 'thinkingTokenBudget' provides the recommended budget for -- token-based providers. module Baikai.ThinkingLevel ( ThinkingLevel (..), @@ -19,24 +19,29 @@ import GHC.Generics (Generic) import Numeric.Natural (Natural) --- | The four buckets pi-mono settled on. Coarse enough that callers --- do not need to learn each provider's TTL conventions, fine enough --- to influence behaviour in a meaningful way. +-- | Provider-neutral reasoning-effort buckets, ordered from least to +-- most effort. Coarse enough that callers do not need to learn each +-- provider's vocabulary, fine enough to influence behaviour in a +-- meaningful way. data ThinkingLevel = ThinkingMinimal | ThinkingLow | ThinkingMedium | ThinkingHigh + | ThinkingXHigh + | ThinkingMax deriving stock (Eq, Ord, Show, Generic) deriving anyclass (FromJSON, ToJSON) --- | The wire string OpenAI's @reasoning_effort@ field expects. +-- | Render the canonical Baikai name for a reasoning-effort level. renderThinkingLevel :: ThinkingLevel -> Text renderThinkingLevel = \case ThinkingMinimal -> "minimal" ThinkingLow -> "low" ThinkingMedium -> "medium" ThinkingHigh -> "high" + ThinkingXHigh -> "xhigh" + ThinkingMax -> "max" -- | Recommended token budget for providers that take an explicit -- count (Anthropic's @thinking.budget_tokens@). @@ -46,3 +51,5 @@ ThinkingLow -> 2048 ThinkingMedium -> 8192 ThinkingHigh -> 16384 + ThinkingXHigh -> 24576 + ThinkingMax -> 32768 diff --git a/test/InteractiveSpec.hs b/test/InteractiveSpec.hs --- a/test/InteractiveSpec.hs +++ b/test/InteractiveSpec.hs @@ -27,6 +27,7 @@ req ^. #extraDirs @?= [] req ^. #safety @?= DefaultSafety req ^. #extraArgs @?= [] + req ^. #effort @?= Nothing providerRenderingTest :: TestTree providerRenderingTest = diff --git a/test/Main.hs b/test/Main.hs --- a/test/Main.hs +++ b/test/Main.hs @@ -27,6 +27,7 @@ import Test.Tasty.HUnit (assertBool, testCase, (@?=)) import Test.Tasty.QuickCheck (Gen) import Test.Tasty.QuickCheck qualified as QC +import ThinkingLevelSpec qualified import TraceSpec qualified import UsageSpec qualified @@ -101,6 +102,7 @@ InteractiveSpec.tests, StreamSpec.tests, SurfaceSpec.tests, + ThinkingLevelSpec.tests, TraceSpec.tests, UsageSpec.tests ] diff --git a/test/ThinkingLevelSpec.hs b/test/ThinkingLevelSpec.hs new file mode 100644 --- /dev/null +++ b/test/ThinkingLevelSpec.hs @@ -0,0 +1,36 @@ +module ThinkingLevelSpec (tests) where + +import Baikai.ThinkingLevel +import Data.Text (Text) +import Test.Tasty (TestTree, testGroup) +import Test.Tasty.HUnit (testCase, (@?=)) + +tests :: TestTree +tests = + testGroup + "ThinkingLevel" + [ testGroup "canonical rendering" renderTests, + testGroup "token budgets" budgetTests + ] + +levels :: [(String, ThinkingLevel, Text, Integer)] +levels = + [ ("minimal", ThinkingMinimal, "minimal", 1024), + ("low", ThinkingLow, "low", 2048), + ("medium", ThinkingMedium, "medium", 8192), + ("high", ThinkingHigh, "high", 16384), + ("xhigh", ThinkingXHigh, "xhigh", 24576), + ("max", ThinkingMax, "max", 32768) + ] + +renderTests :: [TestTree] +renderTests = + [ testCase name $ renderThinkingLevel level @?= expected + | (name, level, expected, _) <- levels + ] + +budgetTests :: [TestTree] +budgetTests = + [ testCase name $ thinkingTokenBudget level @?= fromInteger expected + | (name, level, _, expected) <- levels + ]