Workbench / Online
BUILD YOUR COMPUTE STACK
Start with the model. MuseBoard maps the memory and hardware. Settings are mirrored to the URL, so any configuration can be shared.
Compute workbench
B / Estimated VRAM
Inference42.7GB
Serving: weights + KV cache + runtime overhead. Estimate — not a guarantee.
- Model weights36.4 GB
- KV cache2.7 GB
- Runtime overhead3.6 GB
Memory headroom
Free after estimated load
Utilization
Target ≤ 90% of 64.0 GB
Decode ceiling
Theoretical, batch 1, bandwidth-bound
VRAM usage
42.7 GB / 64.0 GB
Pinned configuration
2 × NVIDIA GeForce RTX 5090
- Total VRAM
- 64.0 GB
- Topology
- Single node · TP
- Interconnect
- PCIe 5.0
- Board power
- 1,150 W
Compatibility / 2 × RTX 5090
- Ready
Inference
42.7 GB · 67% of 2 × RTX 5090
- Limited
Fine-tuning
66.2 GB · Needs 4 × RTX 5090
- Cluster
Training
1,299 GB · Needs 48 GPUs — exceeds one consumer node
Alternative configurations
- 01SERVE / READY
Inference
Estimate memory requirements for serving models.
weights + KV cache + overhead
Open in workbench - 02ADAPT / LORA
Fine-tuning
Explore memory requirements for adapting existing models.
frozen base + adapters + activations
Open in workbench - 03TRAIN / ADAM
Training
Understand large-scale compute requirements.
weights + grads + optimizer + activations
Open in workbench
MuseBoard gives first-order estimates for planning conversations. It does not replace profiling, capacity testing or detailed ML infrastructure planning.