Enum VisionModel 

pub enum VisionModel {
    Claude,
    ClaudeStandard,
    Gpt6,
    Gpt4o,
    Gpt5,
    Gemini15,
    LlamaVision,
    QwenVl,
    DeepseekVl,
    DeepseekFlash,
    KimiVision,
    GenericVision,
}

Supported vision model families with exact or explicitly advisory profiles.

Variants

Claude

Claude 4.7+ high-resolution vision: 28×28 patches, 2576px edge / 4784-token budget.

ClaudeStandard

Earlier Claude vision models: 28×28 patches, 1568px edge / 1568-token budget.

Gpt6

Current OpenAI vision models (GPT-6 / GPT-5.6): 32×32 patches with a 1.2 multiplier.

Gpt4o

GPT-4o / GPT-4.5 high detail: fits in 2048x2048, scales short side to 768, then 512x512 tiles.

Gpt5

Legacy GPT-5/5.1 high detail: 70 base tokens + 140 per 512×512 tile.

Gemini15

Gemini 2.0/3.0: flat 258 tokens if ≤ 384x384, else 258 per 768x768 tile.

LlamaVision

Meta Llama 3.2/3.3 Vision (Mllama): 560×560 tiles, aspect-ratio canvas capped at 4 tiles. (Llama 4 uses a different native-multimodal vision encoder and is not modeled by this arm.)

QwenVl

Alibaba Qwen2-VL / Qwen2.5-VL: 28px effective grid (14px ViT patch × 2×2 merge), token clamp.

DeepseekVl

DeepSeek-VL2: 384×384 base + dynamic 384 local tiles (open-weights; value is local-context savings).

DeepseekFlash

DeepSeek Flash API: current multimodal endpoint, capped at 384 image tokens per image.

KimiVision

Kimi K2.5/K2.6/K3 native vision. Moonshot does not publish a fixed billing grid; estimate is advisory.

GenericVision

Popular open/API vision families without a stable public billing grid. Estimates are advisory; resizing remains useful and deterministic.

Implementations

impl VisionModel

pub fn parse(value: &str) -> Option<Self>

Parse the model aliases accepted by the CLI, MCP server, and Python binding.

pub fn display_name(self) -> &'static str

Trait Implementations

impl Clone for VisionModel

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Copy for VisionModel

impl Debug for VisionModel

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.