renderers 0.1.10.dev10__tar.gz → 0.1.10.dev11__tar.gz

{renderers-0.1.10.dev10 → renderers-0.1.10.dev11}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: renderers
-Version: 0.1.10.dev10
+Version: 0.1.10.dev11
 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`, `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`, `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.
 
 ## API
 
@@ -175,7 +175,7 @@ Fallback for unsupported text-only models. Wraps `apply_chat_template` and accep
 
 ## Roadmap
 
-- **VLM expansion.** `ImagePart` support exists for Qwen3-VL and Qwen3.5-family multimodal templates. Remaining work: video support, broader VLM coverage, and more RL validation.
+- **VLM expansion.** `ImagePart` support exists for Qwen3-VL, Qwen3.5-family, Gemma 4, and Kimi K2.5 / K2.6 multimodal templates. Remaining work: audio/video support, broader VLM coverage, and more RL validation. Gemma 4 image preprocessing requires a Transformers release that provides `Gemma4Processor`.
 - **Patched chat templates.** Some shipped templates re-tokenize history or normalize JSON in ways that break token identity. Plan: a `use_patched` opt-in per renderer that renders the same surface form while avoiding known-bad patterns. (Auto-stripping thinking from past turns is *not* one of these — that's intended template behaviour the renderer reproduces; use `thinking_retention` to override it.)
 
 ## Testing

{renderers-0.1.10.dev10 → renderers-0.1.10.dev11}/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`, `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`, `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.
 
 ## API
 
@@ -159,7 +159,7 @@ Fallback for unsupported text-only models. Wraps `apply_chat_template` and accep
 
 ## Roadmap
 
-- **VLM expansion.** `ImagePart` support exists for Qwen3-VL and Qwen3.5-family multimodal templates. Remaining work: video support, broader VLM coverage, and more RL validation.
+- **VLM expansion.** `ImagePart` support exists for Qwen3-VL, Qwen3.5-family, Gemma 4, and Kimi K2.5 / K2.6 multimodal templates. Remaining work: audio/video support, broader VLM coverage, and more RL validation. Gemma 4 image preprocessing requires a Transformers release that provides `Gemma4Processor`.
 - **Patched chat templates.** Some shipped templates re-tokenize history or normalize JSON in ways that break token identity. Plan: a `use_patched` opt-in per renderer that renders the same surface form while avoiding known-bad patterns. (Auto-stripping thinking from past turns is *not* one of these — that's intended template behaviour the renderer reproduces; use `thinking_retention` to override it.)
 
 ## Testing

{renderers-0.1.10.dev10 → renderers-0.1.10.dev11}/docs/renderer-config.md

@@ -30,6 +30,7 @@ chat-template kwargs. Those fields are covered by parity tests against
 | 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-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` | - |
 | GLM-4.5 | `GLM45RendererConfig` | `enable_thinking` | - |
 | gpt-oss | `GptOssRendererConfig` | `reasoning_effort`, `conversation_start_date` | `use_system_prompt`, `knowledge_cutoff`, `model_identity`, `auto_drop_analysis` |
@@ -131,6 +132,7 @@ 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` |
+| 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` |
 | gpt-oss | `auto_drop_analysis=False -> all`, else `tool_cycle` |

{renderers-0.1.10.dev10 → renderers-0.1.10.dev11}/renderers/__init__.py

@@ -50,6 +50,7 @@ from renderers.configs import (
     GLM45RendererConfig,
     GLM51RendererConfig,
     GLM5RendererConfig,
+    Gemma4RendererConfig,
     GptOssRendererConfig,
     Hy3RendererConfig,
     KimiK25RendererConfig,
@@ -90,6 +91,7 @@ _LAZY_RENDERERS: dict[str, str] = {
     "GLM51Renderer": "renderers.glm5",
     "GLM5Renderer": "renderers.glm5",
     "GptOssRenderer": "renderers.gpt_oss",
+    "Gemma4Renderer": "renderers.gemma4",
     "Hy3Renderer": "renderers.hy3",
     "KimiK25Renderer": "renderers.kimi_k25",
     "KimiK2Renderer": "renderers.kimi_k2",
@@ -142,6 +144,8 @@ __all__ = [
     "GLM51RendererConfig",
     "GLM5Renderer",
     "GLM5RendererConfig",
+    "Gemma4Renderer",
+    "Gemma4RendererConfig",
     "GptOssRenderer",
     "GptOssRendererConfig",
     "Hy3Renderer",

{renderers-0.1.10.dev10 → renderers-0.1.10.dev11}/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.dev10'
-__version_tuple__ = version_tuple = (0, 1, 10, 'dev10')
+__version__ = version = '0.1.10.dev11'
+__version_tuple__ = version_tuple = (0, 1, 10, 'dev11')
 
 __commit_id__ = commit_id = None

{renderers-0.1.10.dev10 → renderers-0.1.10.dev11}/renderers/base.py

@@ -1035,6 +1035,14 @@ MODEL_RENDERER_MAP: dict[str, str] = {
     "Qwen/Qwen3-VL-4B-Instruct": "qwen3-vl",
     "Qwen/Qwen3-VL-8B-Instruct": "qwen3-vl",
     "Qwen/Qwen3-VL-30B-A3B-Instruct": "qwen3-vl",
+    # Gemma 4 instruction checkpoints share Google's canonical turn/tool
+    # grammar and dynamic Gemma4Processor image expansion. E2B/E4B omit the
+    # disabled-thinking empty-channel prefill used by the 26B/31B revision;
+    # Gemma4Renderer detects that small template variant per tokenizer.
+    "google/gemma-4-E2B-it": "gemma4",
+    "google/gemma-4-E4B-it": "gemma4",
+    "google/gemma-4-26B-A4B-it": "gemma4",
+    "google/gemma-4-31B-it": "gemma4",
     # GLM-5 family (GLM-4.7 reuses the GLM-5 template).
     "zai-org/GLM-5": "glm-5",
     "zai-org/GLM-5-FP8": "glm-5",
@@ -1107,6 +1115,10 @@ MULTIMODAL_MODELS: dict[str, set[str]] = {
     "Qwen/Qwen3-VL-4B-Instruct": {"image"},
     "Qwen/Qwen3-VL-8B-Instruct": {"image"},
     "Qwen/Qwen3-VL-30B-A3B-Instruct": {"image"},
+    "google/gemma-4-E2B-it": {"image"},
+    "google/gemma-4-E4B-it": {"image"},
+    "google/gemma-4-26B-A4B-it": {"image"},
+    "google/gemma-4-31B-it": {"image"},
     # Qwen3.5 is itself a VLM family (HF tag ``image-text-to-text``,
     # processor class ``Qwen3VLProcessor``) — same vision tokens and
     # image-processor as Qwen3-VL, with a different tool-call format.
@@ -1326,6 +1338,7 @@ def _populate_registry():
     from renderers.glm5 import GLM5Renderer, GLM51Renderer
     from renderers.glm45 import GLM45Renderer
     from renderers.gpt_oss import GptOssRenderer
+    from renderers.gemma4 import Gemma4Renderer
     from renderers.hy3 import Hy3Renderer
     from renderers.kimi_k2 import KimiK2Renderer
     from renderers.kimi_k25 import KimiK25Renderer
@@ -1354,6 +1367,7 @@ def _populate_registry():
             "qwen3": Qwen3Renderer,
             "prime-qwen3": PrimeQwen3Renderer,
             "qwen3-vl": Qwen3VLRenderer,
+            "gemma4": Gemma4Renderer,
             "qwen3.5": Qwen35Renderer,
             "qwen3.6": Qwen36Renderer,
             "glm-5": GLM5Renderer,

{renderers-0.1.10.dev10 → renderers-0.1.10.dev11}/renderers/client.py

@@ -428,6 +428,7 @@ def _build_mm_features(
     (``MultiModalData``) is already framework-agnostic and does not need
     to change. Don't pre-build the abstraction with one engine in tree.
     """
+    from renderers.gemma4 import Gemma4Renderer
     from renderers.qwen3_vl import Qwen3VLRenderer
     from renderers.qwen35 import Qwen35Renderer
 
@@ -443,6 +444,8 @@ def _build_mm_features(
     # the family default and matches every Qwen-VL processor in tree.
     if issubclass(renderer_cls, (Qwen3VLRenderer, Qwen35Renderer)):
         return _build_qwen_vl_features(mm_data, spatial_merge_size=2)
+    if issubclass(renderer_cls, Gemma4Renderer):
+        return _build_gemma4_features(mm_data)
 
     raise NotImplementedError(
         f"Multimodal serialization not implemented for {renderer_cls.__name__}. "
@@ -450,6 +453,69 @@ def _build_mm_features(
     )
 
 
+def _build_gemma4_features(mm_data: MultiModalData) -> dict[str, Any]:
+    """vLLM features payload for Gemma 4 image inputs.
+
+    Hugging Face names the position field ``image_position_ids`` while
+    vLLM's Gemma 4 processor schema calls it ``pixel_position_ids``. Keep
+    renderer output faithful to the HF processor and translate at this
+    engine-specific boundary.
+    """
+    try:
+        import torch
+        from transformers.feature_extraction_utils import BatchFeature
+        from vllm.entrypoints.scale_out.token_in_token_out.mm_serde import (
+            encode_mm_kwargs_item,
+        )
+        from vllm.multimodal.inputs import (
+            MultiModalFieldConfig,
+            MultiModalKwargsItems,
+        )
+    except ImportError as exc:
+        raise RuntimeError(
+            "Gemma 4 multimodal generate via /inference/v1/generate requires "
+            "a vLLM release with Gemma 4 support and `torch`."
+        ) from exc
+
+    out: dict[str, Any] = {
+        "mm_hashes": {},
+        "mm_placeholders": {},
+        "kwargs_data": {},
+    }
+    image_items = mm_data.mm_items.get("image") or []
+    if image_items:
+        pixel_values = torch.cat(
+            [torch.as_tensor(item["pixel_values"]) for item in image_items], dim=0
+        )
+        pixel_position_ids = torch.cat(
+            [torch.as_tensor(item["image_position_ids"]) for item in image_items],
+            dim=0,
+        )
+        hf_inputs = BatchFeature(
+            data={
+                "pixel_values": pixel_values,
+                "pixel_position_ids": pixel_position_ids,
+            }
+        )
+        field_config = {
+            "pixel_values": MultiModalFieldConfig.batched("image"),
+            "pixel_position_ids": MultiModalFieldConfig.batched("image"),
+        }
+        kwargs_items = MultiModalKwargsItems.from_hf_inputs(hf_inputs, field_config)
+        out["kwargs_data"]["image"] = [
+            encode_mm_kwargs_item(item) for item in kwargs_items["image"]
+        ]
+        out["mm_hashes"]["image"] = list(mm_data.mm_hashes.get("image") or [])
+        out["mm_placeholders"]["image"] = [
+            {"offset": placeholder.offset, "length": placeholder.length}
+            for placeholder in mm_data.mm_placeholders.get("image") or []
+        ]
+
+    if not any(out["kwargs_data"].values()):
+        out["kwargs_data"] = None
+    return out
+
+
 def _build_qwen_vl_features(
     mm_data: MultiModalData, *, spatial_merge_size: int
 ) -> dict[str, Any]:

{renderers-0.1.10.dev10 → renderers-0.1.10.dev11}/renderers/configs.py

@@ -270,6 +270,33 @@ class Qwen3VLRendererConfig(BaseRendererConfig):
     _internal_fields = frozenset({"image_cache_max"})
 
 
+class Gemma4RendererConfig(BaseRendererConfig):
+    """Gemma 4 renderer config."""
+
+    name: Literal["gemma4"] = "gemma4"
+
+    enable_thinking: bool = False
+    """Enable Gemma 4's thinking mode. Mirrors the canonical template kwarg."""
+
+    preserve_thinking: bool = False
+    """Keep thinking on historical tool-call turns when the template permits it."""
+
+    image_cache_max: int = 256
+    """FIFO bound on processed image entries. Renderer-internal."""
+
+    _internal_fields = frozenset({"image_cache_max"})
+
+    @model_validator(mode="after")
+    def _check_thinking_retention(self):
+        _reject_thinking_retention_conflict(
+            self,
+            "preserve_thinking",
+            true_implies="all",
+            false_implies="tool_cycle",
+        )
+        return self
+
+
 class GLM5RendererConfig(BaseRendererConfig):
     """GLM-5 renderer config."""
 
@@ -756,6 +783,7 @@ RendererConfig = Annotated[
         Qwen35RendererConfig,
         Qwen36RendererConfig,
         Qwen3VLRendererConfig,
+        Gemma4RendererConfig,
         GLM5RendererConfig,
         GLM51RendererConfig,
         GLM45RendererConfig,
@@ -799,6 +827,7 @@ _CONFIG_BY_NAME: dict[str, type[BaseRendererConfig]] = {
     "qwen3.5": Qwen35RendererConfig,
     "qwen3.6": Qwen36RendererConfig,
     "qwen3-vl": Qwen3VLRendererConfig,
+    "gemma4": Gemma4RendererConfig,
     "glm-5": GLM5RendererConfig,
     "glm-5.1": GLM51RendererConfig,
     "glm-4.5": GLM45RendererConfig,
@@ -853,6 +882,7 @@ __all__ = [
     "GLM45RendererConfig",
     "GLM51RendererConfig",
     "GLM5RendererConfig",
+    "Gemma4RendererConfig",
     "GptOssRendererConfig",
     "Hy3RendererConfig",
     "KimiK25RendererConfig",