Window: "LM Studio", App: LM Studio.
0 standard window LM Studio, Secondary Actions: Raise
	1 container LM Studio
		2 container
			3 HTML content LM Studio, URL: file:///Applications/LM%20Studio.app/Contents/Resources/app/.webpack/renderer/index.html#/chat
				4 container
					5 button Back
						6 image
					7 text Qwen3 4B
					8 text Estimated Memory Usage
					9 text Beta
					10 text GPU
					11 text 2.80 GB
					12 text Total
					13 text 2.80 GB
					14 container
						15 text Context Length
						16 image
						17 text field (settable) 8192
						18 text Model supports up to
						19 text 40960
						20 text tokens
						21 slider Value: 13.16992, Secondary Actions: Increment, Decrement
						22 text GPU Offload
						23 image
						24 text field (settable) 36
						25 slider Value: 36, Secondary Actions: Increment, Decrement
						26 text CPU Thread Pool Size
						27 image
						28 text field (settable) 10
						29 slider Value: 10, Secondary Actions: Increment, Decrement
						30 text Evaluation Batch Size
						31 image
						32 text field (settable) 2048
						33 text Physical Batch Size
						34 image
						35 text field (settable) 512
						36 text Max Concurrent Predictions
						37 image
						38 text Experimental
						39 button Reset
							40 image
						41 text field (settable) 1
						42 text Unified KV Cache
						43 image
						44 text Experimental
						45 switch Unified KV Cache, Value: on
						46 text Context Checkpoints
						47 image
						48 text field (settable) 32
						49 text Reasoning Budget Message
						50 image
						51 text field (settable) Enter text here
						52 text RoPE Frequency Base
						53 image
						54 checkbox RoPE Frequency Base, Value: 0
						55 text Auto RoPE Frequency Scale
						56 image
						57 checkbox RoPE Frequency Scale, Value: 0
						58 text Auto Offload KV Cache to GPU Memory
						59 image
						60 switch Offload KV Cache to GPU Memory, Value: on
						61 text Keep Model in Memory
						62 image
						63 switch Keep Model in Memory, Value: on
						64 text Try mmap()
						65 image
						66 switch Try mmap(), Value: on
						67 text Seed
						68 image
						69 checkbox Seed, Value: 1
							70 image
						71 button Reset
							72 image
						73 text field (settable) 271828
						74 text Speculative Decoding
						75 image
						76 combo box (settable) Off
							77 text Off
						78 button Show prompt template editor
						79 image
						80 button Show prompt template editor
							81 image
						82 text Flash Attention
						83 image
						84 switch Flash Attention, Value: on
						85 text K Cache Quantization Type
						86 image
						87 checkbox K Cache Quantization Type, Value: 0
						88 text Experimental V Cache Quantization Type
						89 image
						90 checkbox V Cache Quantization Type, Value: 0
						91 text Experimental llama.cpp Arguments Override
						92 switch llama.cpp Arguments Override, Value: off
					93 checkbox Remember settings forqwen3-4b, Value: 1
						94 image
					95 container
						96 text Remember settings for qwen3-4b
					97 switch on
					98 text Show advanced settings
					99 button Cancel
					100 button Load Model ⌘⏎
						101 text Load Model
						102 text ⌘⏎
	103 close button
	104 full screen button Help: this button also has an action to zoom the window, Secondary Actions: zoom the window
	105 minimize button
106 menu bar
	107 LM Studio
	108 File
	109 Edit
	110 View
	111 Window
	112 Help

The focused UI element is 4 container