Can I run it?
Can I run Sharp Spark X2.5 4B on Apple M4 Max (40-core GPU)?
Runs great
Runs great fully in unified memory on your Apple M4 Max (40-core GPU) at Q6_K, 8K context, about 62–85 tok/s. Weights 3.36 GB plus 0.33 GB of KV cache — 4.03 GB of 36.0 GB usable. Up to 512K context still fits.
Best quantQ6_K
Working context8K
Max that fits512K
Speed62–85 tok/sEstimate from memory bandwidth, not a benchmark. Method.
Weights3.36 GB
KV cache0.33 GB
Total4.03 GB
Usable36.0 GB
Sharp Spark X2.5 4B details · Apple M4 Max (40-core GPU) page · How this estimate is made
Run commands
llama.cpp
llama-server -hf peculiar-ragdoll/Sharp-Spark-X2.5-4B-GGUF:Q6_K -c 8192 -ngl 36
No matching name in the Ollama library snapshot.
LM Studio
lms get peculiar-ragdoll/Sharp-Spark-X2.5-4B-GGUF # Then load quant Q6_K
Assumptions
- Estimate from memory bandwidth, not a benchmark. Decode speed uses 55–75% of peak bandwidth.
- KV cache is FP16. Engine overhead assumes llama.cpp (0.3 GB + 1% of the weights). Ollama is closer to 0.9 GB + 3%, LM Studio closer to 1 GB + 3%.
- Q6_K size is the real GGUF file size from Hugging Face.
- Unified memory uses 75% for the GPU by default, and always leaves about 5 GB for the OS.
- Speed is quoted at 8K context, not at the maximum context that still fits.
Larger models that also fit
- Can I run Qwen3.8 27B on Apple M4 Max (40-core GPU)? — Runs, UD-Q8_K_L, 11–14 tok/s (estimate)
- Can I run gemma 4 12b it on Apple M4 Max (40-core GPU)? — Runs, BF16, 11–15 tok/s (estimate)
- Can I run Qwen3.5 9B on Apple M4 Max (40-core GPU)? — Runs great, BF16, 17–23 tok/s (estimate)
Updated 2026-10-24. Profile on this page is the device default.