Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (101 of 101 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (64 of 64 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (26 of 26 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (469 of 469 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (790 of 790 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (95 of 95 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (145 of 145 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (86 of 86 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (62 of 62 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 99.3% (144 of 145 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (1076 of 1076 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (1074 of 1074 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (235 of 235 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (10 of 10 strings)

Translated using Weblate (Norwegian Bokmål)

Currently translated at 100.0% (790 of 790 strings)

Co-authored-by: Hosted Weblate <hosted@weblate.org>
Co-authored-by: OverTheHillsAndFarAway <prosjektx@users.noreply.hosted.weblate.org>
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/common/nb_NO/
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/components-dialog/nb_NO/
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/components-player/nb_NO/
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/config-cameras/nb_NO/
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/config-global/nb_NO/
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/views-configeditor/nb_NO/
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/views-events/nb_NO/
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/views-explore/nb_NO/
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/views-exports/nb_NO/
Translate-URL: https://hosted.weblate.org/projects/frigate-nvr/views-settings/nb_NO/
Translation: Frigate NVR/Config - Cameras
Translation: Frigate NVR/Config - Global
Translation: Frigate NVR/common
Translation: Frigate NVR/components-dialog
Translation: Frigate NVR/components-player
Translation: Frigate NVR/views-configeditor
Translation: Frigate NVR/views-events
Translation: Frigate NVR/views-explore
Translation: Frigate NVR/views-exports
Translation: Frigate NVR/views-settings

frigate/api/chat.py

@@ -36,7 +36,6 @@ from frigate.api.defs.response.chat_response import (
 )
 from frigate.api.defs.tags import Tags
 from frigate.api.event import events
-from frigate.config import FrigateConfig
 from frigate.genai.utils import build_assistant_message_for_conversation
 from frigate.jobs.vlm_watch import (
     get_vlm_watch_job,
@@ -402,38 +401,9 @@ def get_tools() -> JSONResponse:
     return JSONResponse(content={"tools": tools})
 
 
-def _resolve_zones(
-    zones: List[str],
-    config: FrigateConfig,
-    target_cameras: List[str],
-) -> List[str]:
-    """Map zone names to their canonical config keys, case-insensitively.
-
-    LLMs frequently echo a user's casing ("Front Yard") instead of the
-    configured key ("front_yard"). The downstream zone filter is a SQLite GLOB
-    over the JSON-encoded zones column, which is case-sensitive — so an
-    unnormalized name silently returns zero matches. Build a lookup over the
-    relevant cameras' configured zones and substitute when we find a match;
-    unknown names pass through so behavior matches what the model asked for.
-    """
-    if not zones:
-        return zones
-
-    lookup: Dict[str, str] = {}
-    for camera_id in target_cameras:
-        camera_config = config.cameras.get(camera_id)
-        if camera_config is None:
-            continue
-        for zone_name in camera_config.zones.keys():
-            lookup.setdefault(zone_name.lower(), zone_name)
-
-    return [lookup.get(z.lower(), z) for z in zones]
-
-
 async def _execute_search_objects(
     arguments: Dict[str, Any],
     allowed_cameras: List[str],
-    config: FrigateConfig,
 ) -> JSONResponse:
     """
     Execute the search_objects tool.
@@ -467,11 +437,6 @@ async def _execute_search_objects(
     # Convert zones array to comma-separated string if provided
     zones = arguments.get("zones")
     if isinstance(zones, list):
-        camera_arg = arguments.get("camera")
-        target_cameras = (
-            [camera_arg] if camera_arg and camera_arg != "all" else allowed_cameras
-        )
-        zones = _resolve_zones(zones, config, target_cameras)
         zones = ",".join(zones)
     elif zones is None:
         zones = "all"
@@ -563,11 +528,6 @@ async def _execute_find_similar_objects(
     sub_labels = arguments.get("sub_labels")
     zones = arguments.get("zones")
 
-    if zones:
-        zones = _resolve_zones(
-            zones, request.app.frigate_config, cameras or list(allowed_cameras)
-        )
-
     similarity_mode = arguments.get("similarity_mode", "fused")
     if similarity_mode not in ("visual", "semantic", "fused"):
         similarity_mode = "fused"
@@ -695,9 +655,7 @@ async def execute_tool(
     logger.debug(f"Executing tool: {tool_name} with arguments: {arguments}")
 
     if tool_name == "search_objects":
-        return await _execute_search_objects(
-            arguments, allowed_cameras, request.app.frigate_config
-        )
+        return await _execute_search_objects(arguments, allowed_cameras)
 
     if tool_name == "find_similar_objects":
         result = await _execute_find_similar_objects(
@@ -877,9 +835,7 @@ async def _execute_tool_internal(
     This is used by the chat completion endpoint to execute tools.
     """
     if tool_name == "search_objects":
-        response = await _execute_search_objects(
-            arguments, allowed_cameras, request.app.frigate_config
-        )
+        response = await _execute_search_objects(arguments, allowed_cameras)
         try:
             if hasattr(response, "body"):
                 body_str = response.body.decode("utf-8")
@@ -943,9 +899,6 @@ async def _execute_start_camera_watch(
 
     await require_camera_access(camera, request=request)
 
-    if zones:
-        zones = _resolve_zones(zones, config, [camera])
-
     genai_manager = request.app.genai_manager
     chat_client = genai_manager.chat_client
     if chat_client is None or not chat_client.supports_vision:

frigate/data_processing/post/review_descriptions.py

@@ -39,8 +39,6 @@ logger = logging.getLogger(__name__)
 
 RECORDING_BUFFER_EXTENSION_PERCENT = 0.10
 MIN_RECORDING_DURATION = 10
-MAX_IMAGE_TOKENS = 24000
-MAX_FRAMES_PER_SECOND = 2
 
 
 class ReviewDescriptionProcessor(PostProcessorApi):
@@ -62,22 +60,14 @@ class ReviewDescriptionProcessor(PostProcessorApi):
     def calculate_frame_count(
         self,
         camera: str,
-        duration: float,
         image_source: ImageSourceEnum = ImageSourceEnum.preview,
         height: int = 480,
     ) -> int:
-        """Calculate optimal number of frames based on event duration, context size,
-        image source, and resolution.
+        """Calculate optimal number of frames based on context size, image source, and resolution.
 
-        Per-image token cost is asked of the GenAI provider so providers that know
-        their model's true cost (e.g. llama.cpp can probe the loaded mmproj) can
-        diverge from the default ~1-token-per-1250-pixels heuristic. The frame
-        budget is bounded by:
-          - remaining context window after prompt + response reservations
-          - a fixed MAX_IMAGE_TOKENS ceiling
-          - MAX_FRAMES_PER_SECOND x duration, to avoid drowning short events in
-            near-duplicate frames where the model latches onto the redundant middle
-            and skips the start/end action
+        Token usage varies by resolution: larger images (ultra-wide aspect ratios) use more tokens.
+        Estimates ~1 token per 1250 pixels. Targets 98% context utilization with safety margin.
+        Capped at 20 frames.
         """
         client = self.genai_manager.description_client
 
@@ -115,15 +105,14 @@ class ReviewDescriptionProcessor(PostProcessorApi):
                 width = target_width
                 height = int(target_width / aspect_ratio)
 
-        tokens_per_image = client.estimate_image_tokens(width, height)
+        pixels_per_image = width * height
+        tokens_per_image = pixels_per_image / 1250
         prompt_tokens = 3800
         response_tokens = 300
-        context_budget = context_size - prompt_tokens - response_tokens
-        image_token_budget = min(context_budget, MAX_IMAGE_TOKENS)
-        max_frames_by_tokens = int(image_token_budget / tokens_per_image)
-        max_frames_by_duration = int(duration * MAX_FRAMES_PER_SECOND)
-        max_frames = min(max_frames_by_tokens, max_frames_by_duration)
-        return max(max_frames, 3)
+        available_tokens = context_size - prompt_tokens - response_tokens
+        max_frames = int(available_tokens / tokens_per_image)
+
+        return min(max(max_frames, 3), 20)
 
     def process_data(
         self, data: dict[str, Any], data_type: PostProcessDataEnum
@@ -387,9 +376,7 @@ class ReviewDescriptionProcessor(PostProcessorApi):
             all_frames.append(os.path.join(preview_dir, file))
 
         frame_count = len(all_frames)
-        desired_frame_count = self.calculate_frame_count(
-            camera, duration=end_time - start_time
-        )
+        desired_frame_count = self.calculate_frame_count(camera)
 
         if frame_count <= desired_frame_count:
             return all_frames
@@ -413,7 +400,7 @@ class ReviewDescriptionProcessor(PostProcessorApi):
         """Get frames from recordings at specified timestamps."""
         duration = end_time - start_time
         desired_frame_count = self.calculate_frame_count(
-            camera, duration, ImageSourceEnum.recordings, height
+            camera, ImageSourceEnum.recordings, height
         )
 
         # Calculate evenly spaced timestamps throughout the duration

frigate/data_processing/post/types.py

@@ -4,10 +4,6 @@ from pydantic import BaseModel, ConfigDict, Field
 class ReviewMetadata(BaseModel):
     model_config = ConfigDict(extra="ignore", protected_namespaces=())
 
-    observations: list[str] = Field(
-        default_factory=list,
-        description="Chronological list of significant observations from the frames, written before the scene narrative is composed.",
-    )
     title: str = Field(
         description="A short title characterizing what took place and where, under 10 words."
     )

frigate/genai/__init__.py

@@ -163,38 +163,6 @@ Each line represents a detection state, not necessarily unique individuals. The
             if prop is not None:
                 prop.update(hints)
 
-        # observations is a chain-of-thought-by-schema field: forcing the model
-        # to enumerate concrete facts before writing scene/title surfaces details
-        # the narrative would otherwise gloss past (e.g. brief vehicle arrivals
-        # overshadowed by a longer activity). The minItems floor scales with
-        # event duration so longer clips get more observations.
-        observations_prop = schema.get("properties", {}).get("observations")
-        if observations_prop is not None:
-            duration_seconds = float(review_data.get("duration") or 0)
-            min_observations = max(3, round(duration_seconds / 5))
-            max_observations = min_observations + 8
-            observations_prop["description"] = (
-                "Enumerate the significant observations across all frames, in "
-                "chronological order, BEFORE composing the scene narrative. "
-                "Include the very start of the activity — for example, a "
-                "vehicle entering the frame or pulling into the driveway — "
-                "even if it lasts only a few frames and the rest of the clip "
-                "is dominated by a longer activity. Include each arrival, "
-                "departure, motion event, object handled, and notable change "
-                "in position or state. Each item is a single concrete fact "
-                "written as a complete sentence (e.g., 'A blue sedan turns "
-                "from the street into the driveway', 'Nick exits the driver "
-                "side carrying a plant pot'). Do not summarize, interpret, or "
-                "assign meaning here — that belongs in the scene field."
-            )
-            observations_prop["minItems"] = min_observations
-            observations_prop["maxItems"] = max_observations
-            observations_prop["items"] = {"type": "string", "minLength": 20}
-
-            required = schema.setdefault("required", [])
-            if "observations" not in required:
-                required.append("observations")
-
         # OpenAI strict mode requires additionalProperties: false on all objects
         schema["additionalProperties"] = False
 
@@ -388,14 +356,6 @@ Guidelines:
         """Get the context window size for this provider in tokens."""
         return 4096
 
-    def estimate_image_tokens(self, width: int, height: int) -> float:
-        """Estimate prompt tokens consumed by a single image of the given dimensions.
-
-        Default heuristic: ~1 token per 1250 pixels. Providers that can measure or
-        know their model's exact image-token cost should override.
-        """
-        return (width * height) / 1250
-
     def embed(
         self,
         texts: list[str] | None = None,

frigate/genai/llama_cpp.py

@@ -42,8 +42,6 @@ class LlamaCppClient(GenAIClient):
     _supports_vision: bool
     _supports_audio: bool
     _supports_tools: bool
-    _image_token_cache: dict[tuple[int, int], int]
-    _text_baseline_tokens: int | None
 
     def _init_provider(self) -> str | None:
         """Initialize the client and query model metadata from the server."""
@@ -54,8 +52,6 @@ class LlamaCppClient(GenAIClient):
         self._supports_vision = False
         self._supports_audio = False
         self._supports_tools = False
-        self._image_token_cache = {}
-        self._text_baseline_tokens = None
 
         base_url = (
             self.genai_config.base_url.rstrip("/")
@@ -276,91 +272,6 @@ class LlamaCppClient(GenAIClient):
             return self._context_size
         return 4096
 
-    def estimate_image_tokens(self, width: int, height: int) -> float:
-        """Probe the llama.cpp server to learn the model's image-token cost at the
-        requested dimensions.
-
-        llama.cpp's image tokenization is a deterministic function of dimensions and
-        the loaded mmproj, so the result is cached per (width, height) for the
-        lifetime of the process. Falls back to the base pixel heuristic if the
-        server is unreachable or the response is malformed.
-        """
-        if self.provider is None:
-            return super().estimate_image_tokens(width, height)
-
-        cached = self._image_token_cache.get((width, height))
-
-        if cached is not None:
-            return cached
-
-        try:
-            baseline = self._probe_baseline_tokens()
-            with_image = self._probe_image_prompt_tokens(width, height)
-            tokens = max(1, with_image - baseline)
-        except Exception as e:
-            logger.debug(
-                "llama.cpp image-token probe failed for %dx%d (%s); using heuristic",
-                width,
-                height,
-                e,
-            )
-            return super().estimate_image_tokens(width, height)
-
-        self._image_token_cache[(width, height)] = tokens
-        logger.debug(
-            "llama.cpp model '%s' uses ~%d tokens for %dx%d images",
-            self.genai_config.model,
-            tokens,
-            width,
-            height,
-        )
-        return tokens
-
-    def _probe_baseline_tokens(self) -> int:
-        """Return prompt_tokens for a minimal text-only request. Cached after first call."""
-        if self._text_baseline_tokens is not None:
-            return self._text_baseline_tokens
-
-        self._text_baseline_tokens = self._probe_prompt_tokens(
-            [{"type": "text", "text": "."}]
-        )
-        return self._text_baseline_tokens
-
-    def _probe_image_prompt_tokens(self, width: int, height: int) -> int:
-        """Return prompt_tokens for a single synthetic image plus minimal text."""
-        img = Image.new("RGB", (width, height), (128, 128, 128))
-        buf = io.BytesIO()
-        img.save(buf, format="JPEG", quality=60)
-        encoded = base64.b64encode(buf.getvalue()).decode("utf-8")
-        return self._probe_prompt_tokens(
-            [
-                {"type": "text", "text": "."},
-                {
-                    "type": "image_url",
-                    "image_url": {"url": f"data:image/jpeg;base64,{encoded}"},
-                },
-            ]
-        )
-
-    def _probe_prompt_tokens(self, content: list[dict[str, Any]]) -> int:
-        """POST a 1-token chat completion and return reported prompt_tokens.
-
-        Uses a generous timeout to absorb a cold model load on the first probe
-        when the server lazily loads models on demand (e.g. llama-swap).
-        """
-        payload = {
-            "model": self.genai_config.model,
-            "messages": [{"role": "user", "content": content}],
-            "max_tokens": 1,
-        }
-        response = requests.post(
-            f"{self.provider}/v1/chat/completions",
-            json=payload,
-            timeout=60,
-        )
-        response.raise_for_status()
-        return int(response.json()["usage"]["prompt_tokens"])
-
     def _build_payload(
         self,
         messages: list[dict[str, Any]],

web/src/components/chat/ChatMessage.tsx

@@ -155,40 +155,14 @@ export function MessageBubble({
         ) : (
           <div
             className={cn(
+              "[&>*:last-child]:inline",
               !isComplete &&
-                "[&>p:last-child]:inline after:ml-0.5 after:inline-block after:h-4 after:w-2 after:animate-cursor-blink after:rounded-sm after:bg-foreground after:align-middle after:content-['']",
+                "after:ml-0.5 after:inline-block after:h-4 after:w-2 after:animate-cursor-blink after:rounded-sm after:bg-foreground after:align-middle after:content-['']",
             )}
           >
             <ReactMarkdown
               remarkPlugins={[remarkGfm]}
               components={{
-                p: ({ node: _n, ...props }) => (
-                  <p className="my-2 first:mt-0 last:mb-0" {...props} />
-                ),
-                ul: ({ node: _n, ...props }) => (
-                  <ul
-                    className="my-2 list-disc space-y-1 pl-6 first:mt-0 last:mb-0"
-                    {...props}
-                  />
-                ),
-                ol: ({ node: _n, ...props }) => (
-                  <ol
-                    className="my-2 list-decimal space-y-1 pl-6 first:mt-0 last:mb-0"
-                    {...props}
-                  />
-                ),
-                li: ({ node: _n, ...props }) => (
-                  <li className="pl-1" {...props} />
-                ),
-                code: ({ node: _n, className, ...props }) => (
-                  <code
-                    className={cn(
-                      "rounded bg-foreground/10 px-1 py-0.5 font-mono text-sm",
-                      className,
-                    )}
-                    {...props}
-                  />
-                ),
                 table: ({ node: _n, ...props }) => (
                   <table
                     className="my-2 w-full border-collapse border border-border"