Explore benchmarks
Filter by model, hardware, workload and quantization. Every score links to the comparisons and experiments behind it.
Top Engines
| # | Engine | Speed score out of 100 |
|---|---|---|
| 1 | llama.cpp†, an approximate comparison, see details | 89 /100 Score |
| ||
| Not ranked in this view | ||
| Not ranked | ExecuTorch | Too few results |
| Not ranked | Magnitude | Too few results |
| Not ranked | MLX LM | Too few results |
| Not ranked | Ollama | Too few results |
| Not ranked | uzu | Too few results |
| Not ranked | LiteRT-LM | Too few results |
| Not ranked | Rapid-MLX | Too few results |
| Not ranked | ZML | Too few results |
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
The score combines counted comparisons. Select one comparison to see its measured values.
† llama.cpp: approximate comparison. The experiments use different effective KV cache precision: fp32, f16, uzu default. The experiments here use different 4-bit schemes: 8da4w, Q4_K_M, mirai-m. The experiments use different effective KV cache precision: f16, uzu default. The experiments here use different 8-bit schemes: Q8_0, mirai-l. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. The experiments use different effective KV cache precision: litert-lm default, f16, magnitude default, model-dtype, uzu default. The experiments use different effective KV cache precision: f16, magnitude default, model-dtype, zml default. The experiments use different effective KV cache precision: f16, model-dtype. The experiments use different effective KV cache precision: fp32, litert-lm default, f16. The experiments here use different 4-bit schemes: 8da4w, litert-int4, Q4_K_M. The experiments use different effective KV cache precision: litert-lm default, f16. The experiments here use different 8-bit schemes: litert-int8, Q8_0.
ExecuTorch: Results in 6 of 16 comparisons; a rank needs 8.
Magnitude: Results in 6 of 16 comparisons; a rank needs 8.
MLX LM: Results in 6 of 16 comparisons; a rank needs 8.
Ollama: Results in 6 of 16 comparisons; a rank needs 8.
uzu: Results in 6 of 16 comparisons; a rank needs 8.
LiteRT-LM: Results in 5 of 16 comparisons; a rank needs 8.
Rapid-MLX: Results in 4 of 16 comparisons; a rank needs 8.
ZML: Results in 2 of 16 comparisons; a rank needs 8.
Coverage
| Desktop | Mobile | |||
|---|---|---|---|---|
| M5 Max 64 GiB macOS 27 Apple M5 Max | OnePlus 13 16 GB Android 16 Qualcomm Snapdragon 8 Elite (SM8750) | Pixel 8 8 GB Android 16 Google Tensor G3 | iPhone 12 4 GB iOS 27 Apple A14 Bionic | |
| llama.cpp | 888/8 out of 100, in 8 of 8 comparisons | 954/4 out of 100, in 4 of 4 comparisons | 822/2 out of 100, in 2 of 2 comparisons | 882/2 out of 100, in 2 of 2 comparisons |
| uzu | 1004/8 out of 100, in 4 of 8 comparisons | Not measured on this device. | Not measured on this device. | 1002/2 out of 100, in 2 of 2 comparisons |
| ExecuTorch | 2/8Not ranked. Results in 2 of 8 comparisons; a rank needs 4. | 1002/4 out of 100, in 2 of 4 comparisons | 1/2Not ranked. Results in 1 of 2 comparisons; a rank needs 2. | 1/2Not ranked. Results in 1 of 2 comparisons; a rank needs 2. |
| LiteRT-LM | 2/8Not ranked. Results in 2 of 8 comparisons; a rank needs 4. | 872/4 out of 100, in 2 of 4 comparisons | 1/2Not ranked. Results in 1 of 2 comparisons; a rank needs 2. | Not measured on this device. |
| Magnitude | 916/8 out of 100, in 6 of 8 comparisons | Not measured on this device. | Not measured on this device. | Not measured on this device. |
| MLX LM | 986/8 out of 100, in 6 of 8 comparisons | Not measured on this device. | Not measured on this device. | Not measured on this device. |
| Ollama | 906/8 out of 100, in 6 of 8 comparisons | Not measured on this device. | Not measured on this device. | Not measured on this device. |
| Rapid-MLX | 964/8 out of 100, in 4 of 8 comparisons | Not measured on this device. | Not measured on this device. | Not measured on this device. |
| ZML | 2/8Not ranked. Results in 2 of 8 comparisons; a rank needs 4. | Not measured on this device. | Not measured on this device. | Not measured on this device. |
Speed
| Rank | Engine | Prefill |
|---|---|---|
| 1 | llama.cpp†, approximate comparison, see details | 259.7 tok/s |
| 2 | uzu†, approximate comparison, see details | 257.7 tok/s |
| 3 | ExecuTorch†, approximate comparison, see details | 125.9 tok/s |
details
LFM2.5 1.2B Instruct on iPhone 12 · 4 GB · 4-bit · Archive 256 input / 128 output · iOS 27
Medians with the model warmed; model loading is excluded. Each table uses one matched comparison.
llama.cpp: † Approximate comparison. The experiments use different effective KV cache precision: fp32, f16, uzu default. The experiments here use different 4-bit schemes: 8da4w, Q4_K_M, mirai-m.
uzu: † Approximate comparison. The experiments use different effective KV cache precision: fp32, f16, uzu default. The experiments here use different 4-bit schemes: 8da4w, Q4_K_M, mirai-m.
ExecuTorch: † Approximate comparison. The experiments use different effective KV cache precision: fp32, f16, uzu default. The experiments here use different 4-bit schemes: 8da4w, Q4_K_M, mirai-m.
Memory usage by engine
| Rank | Engine | Memory |
|---|---|---|
| Not ranked | ExecuTorch | Not measured |
| Not ranked | llama.cpp | Not measured |
| Not ranked | uzu | Not measured |
details
LFM2.5 1.2B Instruct on iPhone 12 · 4 GB · 4-bit · Archive 256 input / 128 output · iOS 27
Memory ranks require comparable, calibrated measurements. Unavailable measurements remain unranked.
ExecuTorch: No peak memory measured. † Approximate comparison. The experiments use different effective KV cache precision: fp32, f16, uzu default. The experiments here use different 4-bit schemes: 8da4w, Q4_K_M, mirai-m.
llama.cpp: No peak memory measured. † Approximate comparison. The experiments use different effective KV cache precision: fp32, f16, uzu default. The experiments here use different 4-bit schemes: 8da4w, Q4_K_M, mirai-m.
uzu: No peak memory measured. † Approximate comparison. The experiments use different effective KV cache precision: fp32, f16, uzu default. The experiments here use different 4-bit schemes: 8da4w, Q4_K_M, mirai-m.
Every comparison
| Comparison | llama.cpp | ExecuTorch | Magnitude | MLX LM | Ollama | uzu | LiteRT-LM | Rapid-MLX | ZML |
|---|---|---|---|---|---|---|---|---|---|
| LFM2.5 1.2B Instruct on iPhone 12 · 4 GB 4-bit · Archive 256 input / 128 output · iOS 27 | 83†, approximate | 94†, approximate | Not run | Not run | Not run | 100†, approximate | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on iPhone 12 · 4 GB 8-bit · Archive 256 input / 128 output · iOS 27 | 94†, approximate | Not run | Not run | Not run | Not run | 100†, approximate | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB 4-bit · Archive 2048 input / 128 output · macOS 27 | 83†, approximate | 86†, approximate | 85†, approximate | Not counted | 84†, approximate | 100†, approximate | Not counted | Not counted | Not run |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB 4-bit · Archive 256 input / 128 output · macOS 27 | 81†, approximate | 87†, approximate | 82†, approximate | Not counted | 84†, approximate | 100†, approximate | Not counted | Not counted | Not run |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB 8-bit · Archive 2048 input / 128 output · macOS 27 | 87†, approximate | Not run | 89†, approximate | 92†, approximate | 87†, approximate | 100†, approximate | 32†, approximate | 92†, approximate | Not run |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB 8-bit · Archive 256 input / 128 output · macOS 27 | 87†, approximate | Not run | 88†, approximate | 96†, approximate | 87†, approximate | 100†, approximate | 32†, approximate | 97†, approximate | Not run |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB bf16 · Archive 2048 input / 128 output · macOS 27 | 92†, approximate | Not run | 100†, approximate | 99†, approximate | 99†, approximate | Not run | Not run | 98†, approximate | 76†, approximate |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB bf16 · Archive 256 input / 128 output · macOS 27 | 95†, approximate | Not run | 100†, approximate | 100†, approximate | 100†, approximate | Not run | Not run | 99†, approximate | 96†, approximate |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB bf16 · 2048 input / 128 output · macOS 27 | 92†, approximate | Not run | Not run | 100†, approximate | Not run | Not run | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB bf16 · 256 input / 128 output · macOS 27 | 85†, approximate | Not run | Not run | 100†, approximate | Not run | Not run | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on OnePlus 13 · 16 GB 4-bit · Archive 2048 input / 128 output · Android 16 | 99†, approximate | 100†, approximate | Not run | Not run | Not run | Not run | Not counted | Not run | Not run |
| LFM2.5 1.2B Instruct on OnePlus 13 · 16 GB 4-bit · Archive 256 input / 128 output · Android 16 | 83†, approximate | 100†, approximate | Not run | Not run | Not run | Not run | Not counted | Not run | Not run |
| LFM2.5 1.2B Instruct on OnePlus 13 · 16 GB 8-bit · Archive 2048 input / 128 output · Android 16 | 100†, approximate | Not run | Not run | Not run | Not run | Not run | 87†, approximate | Not run | Not run |
| LFM2.5 1.2B Instruct on OnePlus 13 · 16 GB 8-bit · Archive 256 input / 128 output · Android 16 | 100†, approximate | Not run | Not run | Not run | Not run | Not run | 86†, approximate | Not run | Not run |
| LFM2.5 1.2B Instruct on Pixel 8 · 8 GB 4-bit · Archive 256 input / 128 output · Android 16 | 78†, approximate | 100†, approximate | Not run | Not run | Not run | Not run | Not counted | Not run | Not run |
| LFM2.5 1.2B Instruct on Pixel 8 · 8 GB 8-bit · Archive 256 input / 128 output · Android 16 | 87†, approximate | Not run | Not run | Not run | Not run | Not run | 100†, approximate | Not run | Not run |
| LFM2.5 1.2B Instruct on iPhone 12 · 4 GB 4-bit · Archive 2048 input / 128 output · iOS 27 Not counted: fewer than two scored engines | Measured†, approximate | Not counted | Not run | Not run | Not run | Not counted | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on iPhone 12 · 4 GB 8-bit · Archive 2048 input / 128 output · iOS 27 Not counted: fewer than two scored engines | Measured†, approximate | Not run | Not run | Not run | Not run | Not counted | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on iPhone 12 · 4 GB bf16 · Archive 2048 input / 128 output · iOS 27 Not counted: fewer than two scored engines | Unsupported | Not run | Not run | Not run | Not run | Not run | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on iPhone 12 · 4 GB bf16 · Archive 256 input / 128 output · iOS 27 Not counted: fewer than two scored engines | Unsupported | Not run | Not run | Not run | Not run | Not run | Not run | Not run | Not run |
| 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 run | Not run | Not counted | Not run | Not run | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on M5 Max · 64 GiB 4-bit · 256 input / 128 output · macOS 27 Not counted: fewer than two scored engines | Measured†, approximate | Not run | Not run | Not counted | Not run | Not run | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on OnePlus 13 · 16 GB bf16 · Archive 2048 input / 128 output · Android 16 Not counted: fewer than two scored engines | Failed | Not run | Not run | Not run | Not run | Not run | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on OnePlus 13 · 16 GB bf16 · Archive 256 input / 128 output · Android 16 Not counted: fewer than two scored engines | Measured | Not run | Not run | Not run | Not run | Not run | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on Pixel 8 · 8 GB 4-bit · Archive 2048 input / 128 output · Android 16 Not counted: fewer than two scored engines | Not counted | Not counted | Not run | Not run | Not run | Not run | Not counted | Not run | Not run |
| LFM2.5 1.2B Instruct on Pixel 8 · 8 GB 8-bit · Archive 2048 input / 128 output · Android 16 Not counted: fewer than two scored engines | Measured†, approximate | Not run | Not run | Not run | Not run | Not run | Not counted | Not run | Not run |
| LFM2.5 1.2B Instruct on Pixel 8 · 8 GB bf16 · Archive 2048 input / 128 output · Android 16 Not counted: fewer than two scored engines | Failed | Not run | Not run | Not run | Not run | Not run | Not run | Not run | Not run |
| LFM2.5 1.2B Instruct on Pixel 8 · 8 GB bf16 · Archive 256 input / 128 output · Android 16 Not counted: fewer than two scored engines | Failed | Not run | Not run | Not run | Not run | Not run | Not run | Not run | Not run |
Each engine’s Speed score in each matched comparison, out of 100, or why it has none. How rankings work