Model techniques map
Techniquesinference & servingKV cache management

implementation detail · filed under inference & serving

Language-model-only serving mode

Skips the vision encoder and multimodal profiling to free memory for KV caching.

Also called language-model-only, Text-Only.

sources
3
models
2
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

optional 2used 1

Documented in

Evidence

3 spans quoted from the sources, strongest treatment first.

use --language-model-only to skip loading the vision encoder and free up memory for KV cache, and enable Expert Parallelism.

usedinference servingin vLLMvLLM

The following command skips the vision encoder and multimodal profiling to free up memory for additional KV cache

optionalinference servingin Qwen3.5-35B-A3BQwen

The following command skips the vision encoder and multimodal profiling to free up memory for additional KV cache: - vllm serve ... --language-model-only

optionalinference servingin Qwen3.6-35B-A3BQwen

Filed alongside

Other methods under inference & serving :: KV cache management.