Home / Multi-GPU
llama.cpp · homogeneous 2–4 GPUs · not a PC builder
Can this model run across multiple GPUs?
Phase 8 answers memory feasibility for identical cards under llama.cpp. Default split is layer (pipeline parallel: each GPU owns a slice of layers; KV stays with those layers). This is not 2 × VRAM = pool, and it is not a complete PC build.
2× AMD Radeon RX 6950 XT 16 GB
Does not fit on this configuration
- Model
- Mixtral 8x22B Instruct · Q4 · 8K
- Runtime / split
- llama.cpp ·
layer· calculator 1.0.0 - Single-GPU status
- DOES_NOT_FIT · required 91.8 GB vs usable 14.4 GB
- Per card advertised / usable
- 16 GB / 14.4 GB
- Aggregate advertised
- 32.0 GB — capacity, not effective model capacity
- Nominal aggregate usable
- 28.8 GB (N × advertised × 0.90)
- Effective per-GPU peak
- 46.3 GB vs usable 14.4 GB
- Per-GPU split (approx.)
- weights 41.4 GB · KV 0.9 GB · runtime 2.6 GB · safety 1.5 GB
- Why
- Even a 2K context exceeds per-GPU usable VRAM after partition and per-device reserves.
- Topology
- Requires space for 2 discrete GPUs. Exact cooler slot width is not in the RigForAI graph. layer split is pipeline parallel and can run over PCIe. KV stays with the layers on each GPU. NVLink is not treated as a 1× aggregate VRAM pool.
- Performance
- Phase 5 predictions are single-GPU. Multi-GPU tok/s is not published.
- GPU-only current cost
- $2380.00 = 2 × $1190.00 at the current per-card matched price. This does not mean 2 units are in stock, and it is not a complete build cost.MSI GAMING Radeon RX 6950 XT X TRIO 16G AMD Radeon RX 6950XT 16 GB GDDR6 · ASIN B09YHXFD19
- Formula
- per_gpu_peak = ceil(weight_bytes/N) + ceil(kv_bytes/N) + runtime_base + safety_base + weight_fractions×ceil(weight_bytes/N). Compare to usable = advertised×0.90. Not (single_gpu_required × N) and not advertised_vram × N as effective capacity.
AMD Radeon RX 6950 XT 16 GB · Single-GPU can-run · Find a single GPU · Build a machine for this GPU setup
Query combinations are not indexed. Source: llama.cpp multi-GPU documentation.