renderers 0.1.10.dev11__tar.gz → 0.1.10.dev15__tar.gz

{renderers-0.1.10.dev11 → renderers-0.1.10.dev15}/.github/workflows/publish-dev.yml

@@ -62,6 +62,6 @@ jobs:
           path: dist/
 
       - name: Publish to PyPI
-        uses: pypa/gh-action-pypi-publish@cef221092ed1bacb1cc03d23a2d87d1d172e277b # v1.14.0
+        uses: pypa/gh-action-pypi-publish@dc37677b2e1c63e2034f94d8a5b11f265b73ba33 # v1.14.2
         with:
           skip-existing: true

{renderers-0.1.10.dev11 → renderers-0.1.10.dev15}/.github/workflows/publish.yml

@@ -87,4 +87,4 @@ jobs:
           path: dist/
 
       - name: Publish to PyPI
-        uses: pypa/gh-action-pypi-publish@cef221092ed1bacb1cc03d23a2d87d1d172e277b # v1.14.0
+        uses: pypa/gh-action-pypi-publish@dc37677b2e1c63e2034f94d8a5b11f265b73ba33 # v1.14.2

{renderers-0.1.10.dev11 → renderers-0.1.10.dev15}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.5
 Name: renderers
-Version: 0.1.10.dev11
+Version: 0.1.10.dev15
 Summary: Chat template renderers — deterministic message-to-token conversion for LLM training
 License-Expression: Apache-2.0
 License-File: LICENSE
@@ -56,7 +56,7 @@ next_prompt_ids = r.bridge_to_next_turn(
 )
 ```
 
-Hand-coded renderers ship for `qwen3`, `qwen3-vl`, `qwen3.5`, `qwen3.6`, `gemma4`, `glm-5`, `glm-5.1`, `glm-4.5`, `minimax-m2`, `deepseek-v3`, `deepseek-r1`, `kimi-k2`, `kimi-k2.5` / `kimi-k2.6`, `laguna-xs.2`, `laguna-xs-2.1`, `laguna-s-2.1`, `laguna-m.1`, `nemotron-3`, `nemotron-3-ultra`, `nemotron-3.5`, `llama-3`, `gpt-oss`, `hy3`, and `prime-qwen3`. Anything else falls back to `DefaultRenderer`, a generic `apply_chat_template` wrapper.
+Hand-coded renderers ship for `qwen3`, `qwen3-vl`, `qwen3.5`, `qwen3.6`, `qwen3.8`, `gemma4`, `glm-5`, `glm-5.1`, `glm-4.5`, `minimax-m2`, `deepseek-v3`, `deepseek-r1`, `kimi-k2`, `kimi-k2.5` / `kimi-k2.6`, `laguna-xs.2`, `laguna-xs-2.1`, `laguna-s-2.1`, `laguna-m.1`, `nemotron-3`, `nemotron-3-ultra`, `nemotron-3.5`, `llama-3`, `gpt-oss`, `hy3`, `inkling` / `inkling-small`, and `prime-qwen3`. Anything else falls back to `DefaultRenderer`, a generic `apply_chat_template` wrapper. `qwen3-vl`, `qwen3.5`, `qwen3.6`, `qwen3.8`, `gemma4`, `kimi-k2.5` / `kimi-k2.6`, and the Inkling checkpoints are multimodal (Inkling handles both image **and** audio).
 
 ## API
 
@@ -159,7 +159,7 @@ renderer = create_renderer(
 
 Discriminated union: every per-renderer config is a variant of `RendererConfig`, dispatched on the `name` field. Bogus combinations (e.g. `add_vision_id` under `name="qwen3"`) error at construction with a `pydantic.ValidationError`. Downstream pydantic configs (prime-rl orchestrator, verifiers `ClientConfig`) hold a single field typed as `RendererConfig` and inherit the same strict-per-variant validation.
 
-When `chat_template_kwargs` is passed with `config=None` / `AutoRendererConfig`, renderers first resolves the concrete renderer from the model name, then validates those kwargs against that renderer's config. `Auto + unknown model + chat_template_kwargs` fails loudly; use an explicit typed config or explicit `DefaultRendererConfig` for opaque fallback templates.
+When `chat_template_kwargs` is passed with `config=None` / `AutoRendererConfig`, renderers first resolves the concrete renderer from the model name, then validates each key against that renderer's explicit template-kwarg allowlist. Renderer-only fields such as `image_cache_max` must be passed through the typed config instead. `Auto + unknown model + chat_template_kwargs` fails loudly; use an explicit typed config or explicit `DefaultRendererConfig` for opaque fallback templates.
 
 One shared behaviour flag lives on typed renderer configs: `thinking_retention`, an optional bridge-policy override. Leave it unset to derive bridge behaviour from the chat template and its renderer-exposed kwargs.
 

{renderers-0.1.10.dev11 → renderers-0.1.10.dev15}/README.md

@@ -40,7 +40,7 @@ next_prompt_ids = r.bridge_to_next_turn(
 )
 ```
 
-Hand-coded renderers ship for `qwen3`, `qwen3-vl`, `qwen3.5`, `qwen3.6`, `gemma4`, `glm-5`, `glm-5.1`, `glm-4.5`, `minimax-m2`, `deepseek-v3`, `deepseek-r1`, `kimi-k2`, `kimi-k2.5` / `kimi-k2.6`, `laguna-xs.2`, `laguna-xs-2.1`, `laguna-s-2.1`, `laguna-m.1`, `nemotron-3`, `nemotron-3-ultra`, `nemotron-3.5`, `llama-3`, `gpt-oss`, `hy3`, and `prime-qwen3`. Anything else falls back to `DefaultRenderer`, a generic `apply_chat_template` wrapper.
+Hand-coded renderers ship for `qwen3`, `qwen3-vl`, `qwen3.5`, `qwen3.6`, `qwen3.8`, `gemma4`, `glm-5`, `glm-5.1`, `glm-4.5`, `minimax-m2`, `deepseek-v3`, `deepseek-r1`, `kimi-k2`, `kimi-k2.5` / `kimi-k2.6`, `laguna-xs.2`, `laguna-xs-2.1`, `laguna-s-2.1`, `laguna-m.1`, `nemotron-3`, `nemotron-3-ultra`, `nemotron-3.5`, `llama-3`, `gpt-oss`, `hy3`, `inkling` / `inkling-small`, and `prime-qwen3`. Anything else falls back to `DefaultRenderer`, a generic `apply_chat_template` wrapper. `qwen3-vl`, `qwen3.5`, `qwen3.6`, `qwen3.8`, `gemma4`, `kimi-k2.5` / `kimi-k2.6`, and the Inkling checkpoints are multimodal (Inkling handles both image **and** audio).
 
 ## API
 
@@ -143,7 +143,7 @@ renderer = create_renderer(
 
 Discriminated union: every per-renderer config is a variant of `RendererConfig`, dispatched on the `name` field. Bogus combinations (e.g. `add_vision_id` under `name="qwen3"`) error at construction with a `pydantic.ValidationError`. Downstream pydantic configs (prime-rl orchestrator, verifiers `ClientConfig`) hold a single field typed as `RendererConfig` and inherit the same strict-per-variant validation.
 
-When `chat_template_kwargs` is passed with `config=None` / `AutoRendererConfig`, renderers first resolves the concrete renderer from the model name, then validates those kwargs against that renderer's config. `Auto + unknown model + chat_template_kwargs` fails loudly; use an explicit typed config or explicit `DefaultRendererConfig` for opaque fallback templates.
+When `chat_template_kwargs` is passed with `config=None` / `AutoRendererConfig`, renderers first resolves the concrete renderer from the model name, then validates each key against that renderer's explicit template-kwarg allowlist. Renderer-only fields such as `image_cache_max` must be passed through the typed config instead. `Auto + unknown model + chat_template_kwargs` fails loudly; use an explicit typed config or explicit `DefaultRendererConfig` for opaque fallback templates.
 
 One shared behaviour flag lives on typed renderer configs: `thinking_retention`, an optional bridge-policy override. Leave it unset to derive bridge behaviour from the chat template and its renderer-exposed kwargs.
 

{renderers-0.1.10.dev11 → renderers-0.1.10.dev15}/docs/renderer-config.md

@@ -19,8 +19,10 @@ construction.
 
 ## Per-renderer configs
 
-Use `type(config).template_field_names()` to inspect the fields that mirror
-chat-template kwargs. Those fields are covered by parity tests against
+Use `type(config).template_field_names()` to inspect the explicit allowlist of
+fields accepted through `chat_template_kwargs`. Every renderer-specific field
+is classified as either a template field or a renderer-only field at class
+definition time. Template fields are covered by parity tests against
 `apply_chat_template` in `tests/test_renderer_config_parity.py`.
 
 | Renderer | Config class | Template fields | Renderer-only fields |
@@ -29,6 +31,7 @@ chat-template kwargs. Those fields are covered by parity tests against
 | PrimeIntellect Qwen3 | `PrimeQwen3RendererConfig` | - | - |
 | Qwen3.5 | `Qwen35RendererConfig` | `enable_thinking`, `add_vision_id` | `image_cache_max` |
 | Qwen3.6 | `Qwen36RendererConfig` | `enable_thinking`, `add_vision_id`, `preserve_thinking` | `image_cache_max` |
+| Qwen3.8 | `Qwen38RendererConfig` | `enable_thinking`, `add_vision_id`, `preserve_thinking`, `reasoning_effort` | `image_cache_max` |
 | Qwen3-VL | `Qwen3VLRendererConfig` | `add_vision_id` | `image_cache_max` |
 | Gemma 4 | `Gemma4RendererConfig` | `enable_thinking`, `preserve_thinking` | `image_cache_max` |
 | GLM-5 / 5.1 | `GLM5RendererConfig` / `GLM51RendererConfig` | `enable_thinking`, `clear_thinking` | - |
@@ -37,6 +40,7 @@ chat-template kwargs. Those fields are covered by parity tests against
 | Hy3 | `Hy3RendererConfig` | `reasoning_effort`, `preserved_thinking`, `is_training`, `raw_last_assistant`, `fallback_strategy` | - |
 | Kimi K2 | `KimiK2RendererConfig` | - | `enable_thinking` |
 | Kimi K2.5 / 2.6 | `KimiK25RendererConfig` | `thinking` | `image_cache_max` |
+| Inkling / Inkling-Small | `InklingRendererConfig` | `reasoning_effort` | `image_cache_max`, `audio_cache_max` |
 | Laguna XS.2 | `LagunaXS2RendererConfig` | `enable_thinking`, `render_assistant_messages_raw` | - |
 | Laguna M.1 | `LagunaM1RendererConfig` | `enable_thinking`, `render_assistant_messages_raw` | - |
 | Laguna XS-2.1 | `LagunaXS21RendererConfig` | `enable_thinking` | - |
@@ -80,8 +84,10 @@ pool = create_renderer_pool(
 ```
 
 Renderers resolves auto configs before applying `chat_template_kwargs`, so the
-kwargs validate against the concrete renderer config. Unknown kwargs, or kwargs
-that conflict with an explicit `thinking_retention`, fail at construction.
+kwargs validate against the concrete renderer's template-field allowlist.
+Unknown kwargs, renderer-only fields, or kwargs that conflict with an explicit
+`thinking_retention` fail at construction. Renderer-only options remain valid
+when supplied through the typed config itself.
 
 Auto-resolution fails loudly for VLMs without an exact registered renderer.
 Text-only unknown models fall back to `DefaultRenderer`, unless
@@ -90,7 +96,10 @@ cannot implement selective bridge retention, so that combination raises.
 `AutoRendererConfig` with `chat_template_kwargs` also raises for unknown models,
 because renderers cannot validate those kwargs without a concrete renderer.
 Use an explicit model-specific config, or `DefaultRendererConfig(...)` when you
-intentionally want opaque `apply_chat_template` kwargs.
+intentionally want opaque `apply_chat_template` kwargs. Even for the default
+renderer, typed fields such as `tool_parser`, `reasoning_parser`, and
+`thinking_retention` must be passed through the config rather than through the
+opaque kwargs mapping.
 
 ## `thinking_retention`
 
@@ -132,6 +141,7 @@ the knobs its template actually exposes:
 | Qwen3 | `enable_thinking=False -> all`, else `tool_cycle` |
 | Qwen3.5 | `enable_thinking=False -> all`, else `tool_cycle` |
 | Qwen3.6 | `preserve_thinking=True -> all`; else `enable_thinking=False -> all`; else `tool_cycle` |
+| Qwen3.8 | `preserve_thinking=True -> all`; else `enable_thinking=False -> all`; else `tool_cycle` |
 | Gemma 4 | `preserve_thinking=True -> all`; else `enable_thinking=False -> all`; else `tool_cycle` |
 | GLM-5 / 5.1 | `clear_thinking=False -> all`; else `enable_thinking=False -> all`; else `tool_cycle` |
 | GLM-4.5 | `enable_thinking=False -> all`, else `tool_cycle` |
@@ -141,7 +151,7 @@ the knobs its template actually exposes:
 | Nemotron-3 / 3.5 | `truncate_history_thinking=False -> all`; else `enable_thinking=False -> all`; else `tool_cycle` |
 | DeepSeek R1 | `template` |
 | MiniMax M2 | `tool_cycle` |
-| DeepSeek V3, Qwen3-VL, Kimi K2, Laguna XS.2 / M.1 / XS-2.1 / S-2.1, Llama 3 | `all` |
+| DeepSeek V3, Qwen3-VL, Kimi K2, Laguna XS.2 / M.1 / XS-2.1 / S-2.1, Llama 3, Inkling | `all` |
 | PrimeIntellect Qwen3 | `all` |
 
 Config construction raises when an explicit template knob directly contradicts

{renderers-0.1.10.dev11 → renderers-0.1.10.dev15}/pyproject.toml

@@ -19,6 +19,10 @@ dependencies = [
     "openai>=1.108.1",
     "tiktoken",
     "jinja2",
+    # Keep this floor compatible with prime-rl's transformers pin. Inkling's
+    # tokenizer and text-only renderer work on older releases; image/audio
+    # processing fails lazily with an upgrade message when InklingProcessor is
+    # unavailable (native support starts in transformers 5.14).
     "transformers>=4.50.0",
     # Used by GptOssRenderer to render and parse harmony tokens. Vendoring
     # OpenAI's reference implementation keeps us byte-identical with vLLM

{renderers-0.1.10.dev11 → renderers-0.1.10.dev15}/renderers/__init__.py

@@ -53,6 +53,7 @@ from renderers.configs import (
     Gemma4RendererConfig,
     GptOssRendererConfig,
     Hy3RendererConfig,
+    InklingRendererConfig,
     KimiK25RendererConfig,
     KimiK2RendererConfig,
     LagunaM1RendererConfig,
@@ -67,6 +68,7 @@ from renderers.configs import (
     PrimeQwen3RendererConfig,
     Qwen35RendererConfig,
     Qwen36RendererConfig,
+    Qwen38RendererConfig,
     Qwen3RendererConfig,
     Qwen3VLRendererConfig,
     RendererConfig,
@@ -93,6 +95,7 @@ _LAZY_RENDERERS: dict[str, str] = {
     "GptOssRenderer": "renderers.gpt_oss",
     "Gemma4Renderer": "renderers.gemma4",
     "Hy3Renderer": "renderers.hy3",
+    "InklingRenderer": "renderers.inkling",
     "KimiK25Renderer": "renderers.kimi_k25",
     "KimiK2Renderer": "renderers.kimi_k2",
     "LagunaM1Renderer": "renderers.laguna_xs2",
@@ -107,6 +110,7 @@ _LAZY_RENDERERS: dict[str, str] = {
     "PrimeQwen3Renderer": "renderers.prime_qwen3",
     "Qwen35Renderer": "renderers.qwen35",
     "Qwen36Renderer": "renderers.qwen36",
+    "Qwen38Renderer": "renderers.qwen38",
     "Qwen3Renderer": "renderers.qwen3",
     "Qwen3VLRenderer": "renderers.qwen3_vl",
 }
@@ -151,6 +155,8 @@ __all__ = [
     "Hy3Renderer",
     "Hy3RendererConfig",
     "ImagePart",
+    "InklingRenderer",
+    "InklingRendererConfig",
     "KimiK25Renderer",
     "KimiK25RendererConfig",
     "KimiK2Renderer",
@@ -188,6 +194,8 @@ __all__ = [
     "Qwen35RendererConfig",
     "Qwen36Renderer",
     "Qwen36RendererConfig",
+    "Qwen38Renderer",
+    "Qwen38RendererConfig",
     "Qwen3Renderer",
     "Qwen3RendererConfig",
     "Qwen3VLRenderer",

{renderers-0.1.10.dev11 → renderers-0.1.10.dev15}/renderers/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '0.1.10.dev11'
-__version_tuple__ = version_tuple = (0, 1, 10, 'dev11')
+__version__ = version = '0.1.10.dev15'
+__version_tuple__ = version_tuple = (0, 1, 10, 'dev15')
 
 __commit_id__ = commit_id = None

{renderers-0.1.10.dev11 → renderers-0.1.10.dev15}/renderers/base.py

@@ -794,10 +794,10 @@ class MultimodalRenderer(Renderer, Protocol):
     """A :class:`Renderer` that supports multimodal inputs (images, video).
 
     Concrete classes (``Qwen3VLRenderer``, ``Qwen35Renderer``,
-    ``Qwen36Renderer``, ``KimiK25Renderer``) implement this Protocol
-    structurally — no explicit inheritance required. Callers that need
-    to drive vLLM's ``multi_modal_data`` features field or carry images
-    forward across turns should dispatch on ``isinstance(r,
+    ``Qwen36Renderer``, ``Qwen38Renderer``, ``KimiK25Renderer``) implement
+    this Protocol structurally — no explicit inheritance required.
+    Callers that need to drive vLLM's ``multi_modal_data`` features field or
+    carry images forward across turns should dispatch on ``isinstance(r,
     MultimodalRenderer)`` and use the extended ``bridge_to_next_turn``
     signature below.
     """
@@ -1031,6 +1031,8 @@ MODEL_RENDERER_MAP: dict[str, str] = {
     "Qwen/Qwen3.5-397B-A17B": "qwen3.5",
     # Qwen3.6.
     "Qwen/Qwen3.6-35B-A3B": "qwen3.6",
+    # Qwen3.8.
+    "Qwen/Qwen3.8-27B": "qwen3.8",
     # Qwen3-VL.
     "Qwen/Qwen3-VL-4B-Instruct": "qwen3-vl",
     "Qwen/Qwen3-VL-8B-Instruct": "qwen3-vl",
@@ -1093,6 +1095,10 @@ MODEL_RENDERER_MAP: dict[str, str] = {
     # GPT-OSS.
     "openai/gpt-oss-20b": "gpt-oss",
     "openai/gpt-oss-120b": "gpt-oss",
+    # Thinking Machines Inkling checkpoints share byte-identical tokenizer,
+    # chat-template, and processor assets (vision + audio; transformers >= 5.14).
+    "thinkingmachines/Inkling": "inkling",
+    "thinkingmachines/Inkling-Small": "inkling",
     # Tencent Hunyuan Hy3 (295B-A21B MoE). The FP8 checkpoint shares the same
     # tokenizer and chat template. Hy3-preview is deliberately unmapped: it
     # ships an older, incompatible template (un-suffixed special tokens,
@@ -1132,6 +1138,8 @@ MULTIMODAL_MODELS: dict[str, set[str]] = {
     # Qwen3.6 extends Qwen3.5's chat template; same VL bits, only
     # tool-call argument serialization differs.
     "Qwen/Qwen3.6-35B-A3B": {"image"},
+    # Qwen3.8 adds reasoning-effort control and preserves thinking by default.
+    "Qwen/Qwen3.8-27B": {"image"},
     # Kimi K2.5 / K2.6 are unified VLMs (HF tag ``image-text-to-text``)
     # with custom processor (``KimiK25Processor`` + ``KimiK25VisionProcessor``).
     # Vision wrap is different from Qwen-VL:
@@ -1141,6 +1149,8 @@ MULTIMODAL_MODELS: dict[str, set[str]] = {
     # ``grid_thws``.
     "moonshotai/Kimi-K2.5": {"image"},
     "moonshotai/Kimi-K2.6": {"image"},
+    "thinkingmachines/Inkling": {"image", "audio"},
+    "thinkingmachines/Inkling-Small": {"image", "audio"},
 }
 
 
@@ -1340,6 +1350,7 @@ def _populate_registry():
     from renderers.gpt_oss import GptOssRenderer
     from renderers.gemma4 import Gemma4Renderer
     from renderers.hy3 import Hy3Renderer
+    from renderers.inkling import InklingRenderer
     from renderers.kimi_k2 import KimiK2Renderer
     from renderers.kimi_k25 import KimiK25Renderer
     from renderers.laguna_s21 import LagunaS21Renderer
@@ -1360,6 +1371,7 @@ def _populate_registry():
     from renderers.qwen3_vl import Qwen3VLRenderer
     from renderers.qwen35 import Qwen35Renderer
     from renderers.qwen36 import Qwen36Renderer
+    from renderers.qwen38 import Qwen38Renderer
 
     RENDERER_REGISTRY.update(
         {
@@ -1370,6 +1382,7 @@ def _populate_registry():
             "gemma4": Gemma4Renderer,
             "qwen3.5": Qwen35Renderer,
             "qwen3.6": Qwen36Renderer,
+            "qwen3.8": Qwen38Renderer,
             "glm-5": GLM5Renderer,
             "glm-5.1": GLM51Renderer,
             "glm-4.5": GLM45Renderer,
@@ -1377,6 +1390,7 @@ def _populate_registry():
             "deepseek-v3": DeepSeekV3Renderer,
             "deepseek-r1": DeepSeekR1Renderer,
             "hy3": Hy3Renderer,
+            "inkling": InklingRenderer,
             "kimi-k2": KimiK2Renderer,
             "kimi-k2.5": KimiK25Renderer,
             "laguna-xs.2": LagunaXS2Renderer,
@@ -1483,12 +1497,31 @@ def _merge_chat_template_kwargs(
         return config
     if not isinstance(chat_template_kwargs, Mapping):
         raise TypeError("chat_template_kwargs must be a mapping.")
+    kwargs = dict(chat_template_kwargs)
+    config_cls = type(config)
+    allowed = config_cls.template_field_names()
+    if config_cls._allow_opaque_template_kwargs:
+        reserved = frozenset(config_cls.model_fields) - allowed - {"name"}
+        unsupported = frozenset(kwargs) & reserved
+    else:
+        unsupported = frozenset(kwargs) - allowed
+    if unsupported:
+        allowed_text = (
+            "opaque Jinja kwargs"
+            if config_cls._allow_opaque_template_kwargs
+            else ", ".join(sorted(allowed)) or "(none)"
+        )
+        raise ValueError(
+            f"Unsupported chat_template_kwargs for {config.name!r}: "
+            f"{sorted(unsupported)}. Allowed: {allowed_text}. Pass "
+            "renderer-internal options through the typed config instead."
+        )
     data: dict[str, Any] = {"name": config.name}
     for field_name in config.__pydantic_fields_set__:
         data[field_name] = getattr(config, field_name)
     data.update(getattr(config, "model_extra", None) or {})
-    data.update(dict(chat_template_kwargs))
-    return type(config).model_validate(data)
+    data.update(kwargs)
+    return config_cls.model_validate(data)
 
 
 def _resolve_renderer_config(