# Compiler OOM on DGX Spark GB10 / Unified Memory machines even for 0.5b size models

**URL:** <https://forum.modular.com/t/compiler-oom-on-dgx-spark-gb10-unified-memory-machines-even-for-0-5b-size-models/3264>\
**Category:** General\
**Tags:** mojo-compiler\
**Created:** [June 21, 2026, 11:06pm UTC](https://forum.modular.com/t/compiler-oom-on-dgx-spark-gb10-unified-memory-machines-even-for-0-5b-size-models/3264 "2026-06-21T23:06:39Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![gchauhan](https://avatars.discourse-cdn.com/v4/letter/g/b5e925/32.png) [@gchauhan](https://forum.modular.com/u/gchauhan)\
**Post date:** [June 21, 2026, 11:06pm UTC](https://forum.modular.com/t/compiler-oom-on-dgx-spark-gb10-unified-memory-machines-even-for-0-5b-size-models/3264/1 "2026-06-21T23:06:39Z")

</div>

On DGX Spark GB10 and Unified memory machines, the default settings for MODULAR\_DEVICE\_CONTEXT\_MEMORY\_MANAGER\_SIZE\_PERCENT=90 result in compiler OOMing even for very small models like Qwen/Qwen2.5-0.5B-Instruct.

After spending a whole day trying various debug settings, code paths, finally figured out the following works.

Change to a lower value like export MODULAR\_DEVICE\_CONTEXT\_MEMORY\_MANAGER\_SIZE\_PERCENT=15 to get things to work. Memory spike dropped from 108G → 33G and model loaded with correct output in 96s.

Working example:

`docker run -d --gpus=all \ -e HF_HUB_OFFLINE=1 -e HF_HOME=/hf \ -e MODULAR_DEVICE_CONTEXT_MEMORY_MANAGER_SIZE_PERCENT=15 \ -v ~/.cache/huggingface:/hf -p 8000:8000 \ ``docker.modular.com/modular/max-nvidia-full:nightly`` \`  
` --model Qwen/Qwen2.5-0.5B-Instruct --devices gpu:0 \`  
` --max-length 2048 --max-batch-size 1`

Hope it helps and saves you time from chasing down different rabbit holes.