---
title: "LocalVlmSettings"
canonical_url: "https://www.nutrient.io/api/python/settings/vision/advanced/local-vlm-settings/"
md_url: "https://www.nutrient.io/api/python/settings/vision/advanced/local-vlm-settings.md"
last_updated: "2026-09-02T09:34:35.870Z"
description: "Settings for LocalVlm. Values fall back through three levels: document → SDK → built-in default."
---

Settings for LocalVlm. Values fall back through three levels: document → SDK → built-in default. Writes target the document only when set on a document's settings, otherwise the SDK globally when set on SdkSettings.

**Tags:** `Vision`, `Advanced`

```python

from nutrient_sdk import LocalVlmSettings

```

## Construction

`LocalVlmSettings` is accessed through a [`Document`](/api/python/document/) instance for per-document overrides, or via [`SdkSettings`](/api/python/settings/document/sdk-settings/) for SDK-wide defaults.

```python

# Per-document override

with Document.open("input.pdf") as doc:
    settings = doc.settings.local_vlm_settings
    settings.some_field = new_value          # mutate fields directly

# SDK-wide default (applies to all documents)

SdkSettings.local_vlm_settings.some_field = new_value

```

Settings are configured by writing to fields on the returned object. The settings property itself cannot be reassigned — `doc.settings.local_vlm_settings = other_settings` is rejected.

## Properties

### batch_size

```python

@property
def batch_size(self) -> int

@batch_size.setter
def batch_size(self, value: int) -> None

```

Number of zone images to include per VLM call.

**Type:** `int`

**Default:** `8`

---

### classification_strategy

```python

@property
def classification_strategy(self) -> VlmClassificationStrategy

@classification_strategy.setter
def classification_strategy(self, value: VlmClassificationStrategy) -> None

```

Strategy for how zones are sent to the VLM for classification.

**Type:** [`VlmClassificationStrategy`](/api/python/enums/vlm-classification-strategy/)

**Default:** `VlmClassificationStrategy.MultiZone`

---

### context_tokens

```python

@property
def context_tokens(self) -> int

@context_tokens.setter
def context_tokens(self, value: int) -> None

```

Context size in tokens; 0 uses the engine default (4096).

**Type:** `int`

**Default:** `8192`

---

### gpu_layers

```python

@property
def gpu_layers(self) -> int

@gpu_layers.setter
def gpu_layers(self, value: int) -> None

```

Model layers to offload to GPU; 0 is pure CPU. GPU backends are runtime-provisioned (NAT-519), never linked.

**Type:** `int`

**Default:** `0`

---

### image_min_tokens

```python

@property
def image_min_tokens(self) -> int

@image_min_tokens.setter
def image_min_tokens(self, value: int) -> None

```

Minimum image tokens for dynamic-resolution vision models; 0 uses the model default. Qwen-VL grounding tasks want at least 1024.

**Type:** `int`

**Default:** `1024`

---

### max_concurrency

```python

@property
def max_concurrency(self) -> int

@max_concurrency.setter
def max_concurrency(self, value: int) -> None

```

Maximum number of concurrent VLM calls. Sizes the in-process context pool: the model weights are shared, each extra context costs only its KV/compute buffers and runs one more call in parallel (callers beyond the pool block for a free context).

**Type:** `int`

**Default:** `1`

---

### max_tokens

```python

@property
def max_tokens(self) -> int

@max_tokens.setter
def max_tokens(self, value: int) -> None

```

Maximum tokens in response (-1 = engine default)

**Type:** `int`

**Default:** `-1`

---

### mmproj_path

```python

@property
def mmproj_path(self) -> str

@mmproj_path.setter
def mmproj_path(self, value: str) -> None

```

Filesystem path to the GGUF multimodal projector (mmproj) matching the model.

**Type:** `str`

**Default:** `string.Empty`

---

### model_path

```python

@property
def model_path(self) -> str

@model_path.setter
def model_path(self, value: str) -> None

```

Filesystem path to the GGUF text model (e.g. a Qwen VLM). Model provisioning through the resource downloader is a follow-up; until then the path must point at an existing file.

**Type:** `str`

**Default:** `string.Empty`

---

### send_full_page_reference

```python

@property
def send_full_page_reference(self) -> bool

@send_full_page_reference.setter
def send_full_page_reference(self, value: bool) -> None

```

Whether to send the full page image as reference context alongside cropped zones.

**Type:** `bool`

**Default:** `false`

---

### system_prompt

```python

@property
def system_prompt(self) -> str

@system_prompt.setter
def system_prompt(self, value: str) -> None

```

Optional custom system prompt for zone classification. When empty, the module uses a default prompt based on the ClassificationStrategy.

**Type:** `str`

**Default:** `string.Empty`

---

### temperature

```python

@property
def temperature(self) -> float

@temperature.setter
def temperature(self, value: float) -> None

```

Temperature for response generation (0.0 = deterministic, 1.0 = creative). Defaults to 0.0 because zone classification and verbatim text extraction are deterministic tasks — randomness only adds run-to-run noise.

**Type:** `float`

**Default:** `0.0`

---

### threads

```python

@property
def threads(self) -> int

@threads.setter
def threads(self, value: int) -> None

```

Inference threads; 0 uses the engine default.

**Type:** `int`

**Default:** `0`

---

---

## Related pages

- [Advanced](/api/python/settings/vision/advanced.md)
- [Per-document override](/api/python/settings/vision/advanced/ai-augmenter-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/ai-processing-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/anthropic-api-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/claude-api-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/content-extraction-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/custom-vlm-api-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/deskew-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/document-layout-json-export-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/finalizer-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/form-labeling-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/form-recognition-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/google-api-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/handwriting-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/inference-layout-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/ocr-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/omr-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/open-ai-api-endpoint-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/open-ai-languages-detection-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/open-ai-picture-alt-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/preprocessing-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/reading-order-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/segmenter-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/table-recognition-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/vision-acceleration-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/vision-descriptor-settings.md)
- [Per-document override](/api/python/settings/vision/advanced/words-detection-settings.md)