Explore benchmarks
Filter by model, hardware, workload and quantization. Every score links to the comparisons and experiments behind it.
Top Engines
Overall needs mobile results; each platform is ranked on its own until then.
Desktop
| # | Engine | Speed score out of 100 |
|---|---|---|
| 1 | MLX LM†, an approximate comparison, see details | 100 /100 Score |
| 2 | llama.cpp†, an approximate comparison, see details | 92 /100 Score |
details
Speed Decode speed: tokens generated per second.
Score: 100 is the best result in every comparison an engine entered; 80 is 80% of the best, on average. Only complete experiments with a passed output check count. How rankings work
† MLX LM: approximate comparison. The experiments use different effective KV cache precision: f16, model-dtype.
† llama.cpp: approximate comparison. The experiments use different effective KV cache precision: f16, model-dtype.
Coverage
Speed
| Rank | Engine | Decode and TTFT |
|---|---|---|
| 1 | MLX LM†, approximate comparison, see details | 189.9 tok/s 87.8 ms TTFT |
| 2 | llama.cpp†, approximate comparison, see details | 175.4 tok/s 133 ms TTFT |
details
LFM2.5 1.2B Instruct on M5 Max · 64 GiB · bf16 · 2048 input / 128 output · macOS 27
Medians with the model warmed; model loading is excluded. Each table uses one matched comparison.
MLX LM: † Approximate comparison. The experiments use different effective KV cache precision: f16, model-dtype.
llama.cpp: † Approximate comparison. The experiments use different effective KV cache precision: f16, model-dtype.
Memory usage by engine
| Rank | Engine | Memory |
|---|---|---|
| Not ranked | llama.cpp | Not measured |
| Not ranked | MLX LM | Not measured |
details
LFM2.5 1.2B Instruct on M5 Max · 64 GiB · bf16 · 2048 input / 128 output · macOS 27
Memory ranks require comparable, calibrated measurements. Unavailable measurements remain unranked.
llama.cpp: No peak memory measured. † Approximate comparison. The experiments use different effective KV cache precision: f16, model-dtype.
MLX LM: No peak memory measured. † Approximate comparison. The experiments use different effective KV cache precision: f16, model-dtype.
Every comparison
| Comparison | llama.cpp | MLX LM |
|---|---|---|
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB bf16 · 2048 input / 128 output · macOS 27 | 92†, approximate | 100†, approximate |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB 4-bit · 2048 input / 128 output · macOS 27 Not counted: fewer than two scored engines | Not counted | Not counted |
Each engine’s Speed score in each matched comparison, out of 100, or why it has none. How rankings work