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 | MLX LM†, an approximate comparison, see details | 100 /100 Score |
| ||
| 2 | Rapid-MLX†, an approximate comparison, see details | 98 /100 Score |
| ||
| 3 | uzu†, an approximate comparison, see details | 94 /100 Score |
| ||
| 4 | Magnitude†, an approximate comparison, see details | 89 /100 Score |
| ||
| 5 | Ollama†, an approximate comparison, see details | 76 /100 Score |
| ||
| 6 | llama.cpp†, an approximate comparison, see details | 72 /100 Score |
| ||
| Not ranked in this view | ||
| Not ranked | LiteRT-LM | Too few results |
| Not ranked | ZML | Too few results |
| Not ranked | ExecuTorch | Not counted |
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.
† MLX LM: approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mlx-affine-4bit-g64, where most use Q4_K_M. The experiments use different effective KV cache precision: litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 8-bit in another scheme: mlx-affine-8bit-g64, where most use Q8_0. The experiments use different effective KV cache precision: f16, magnitude default, model-dtype, zml default.
† Rapid-MLX: approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mlx-affine-4bit-g64, where most use Q4_K_M. The experiments use different effective KV cache precision: litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 8-bit in another scheme: mlx-affine-8bit-g64, where most use Q8_0. The experiments use different effective KV cache precision: f16, magnitude default, model-dtype, zml default.
† uzu: approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mirai-m, where most use Q4_K_M. The experiments use different effective KV cache precision: litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 8-bit in another scheme: mirai-l, where most use Q8_0.
† Magnitude: approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its phases were timed by the harness clock around the engine API, so they include transport that engine counters leave out. 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.
† Ollama: approximate comparison. 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.
† llama.cpp: approximate comparison. 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.
LiteRT-LM: Results in 1 of 3 comparisons; a rank needs 2.
ZML: Results in 1 of 3 comparisons; a rank needs 2.
ExecuTorch: Its output check failed: 6/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128.
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 | 723/3 out of 100, in 3 of 3 comparisons | 1001/1 out of 100, in 1 of 1 comparisons | 1001/1 out of 100, in 1 of 1 comparisons | 532/2 out of 100, in 2 of 2 comparisons |
| LiteRT-LM | 1/3Not ranked. Results in 1 of 3 comparisons; a rank needs 2. | 641/1 out of 100, in 1 of 1 comparisons | 931/1 out of 100, in 1 of 1 comparisons | Not measured on this device. |
| uzu | 942/3 out of 100, in 2 of 3 comparisons | Not measured on this device. | Not measured on this device. | 1002/2 out of 100, in 2 of 2 comparisons |
| Magnitude | 893/3 out of 100, in 3 of 3 comparisons | Not measured on this device. | Not measured on this device. | Not measured on this device. |
| MLX LM | 1003/3 out of 100, in 3 of 3 comparisons | Not measured on this device. | Not measured on this device. | Not measured on this device. |
| Ollama | 763/3 out of 100, in 3 of 3 comparisons | Not measured on this device. | Not measured on this device. | Not measured on this device. |
| Rapid-MLX | 983/3 out of 100, in 3 of 3 comparisons | Not measured on this device. | Not measured on this device. | Not measured on this device. |
| ExecuTorch | Not counted Its output check failed: 6/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128. | Not counted Its output check failed: 5/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128. | Not counted Its output check failed: 5/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128. | Not counted Its output check failed: 5/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128. |
| ZML | 1/3Not ranked. Results in 1 of 3 comparisons; a rank needs 2. | Not measured on this device. | Not measured on this device. | Not measured on this device. |
Speed
| Rank | Engine | Decode and TTFT |
|---|---|---|
| 1 | Rapid-MLX†, approximate comparison, see details | 1,108 tok/s 24.5 ms TTFT |
| 2 | MLX LM†, approximate comparison, see details | 1,102 tok/s 23.2 ms TTFT |
| 3 | uzu†, approximate comparison, see details | 1,001 tok/s 67.9 ms TTFT |
| 4 | Magnitude†, approximate comparison, see details | 911.9 tok/s 18.7 ms TTFT |
| 5 | Ollama†, approximate comparison, see details | 742.8 tok/s 21.4 ms TTFT |
| 6 | llama.cpp†, approximate comparison, see details | 684.9 tok/s 23.2 ms TTFT |
| Not ranked | ExecuTorch | Not counted |
| Not ranked | LiteRT-LM | Not counted |
details
LFM2.5 350M on M5 Max · 64 GiB · 4-bit · Archive 256 input / 128 output · macOS 27
Medians with the model warmed; model loading is excluded. Each table uses one matched comparison.
Rapid-MLX: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mlx-affine-4bit-g64, where most use Q4_K_M.
MLX LM: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mlx-affine-4bit-g64, where most use Q4_K_M.
uzu: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mirai-m, where most use Q4_K_M.
Magnitude: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its phases were timed by the harness clock around the engine API, so they include transport that engine counters leave out.
Ollama: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default.
llama.cpp: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default.
ExecuTorch: Its output check failed: 6/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128. † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mlx-4w-g64, where most use Q4_K_M.
LiteRT-LM: Its output check failed: 6/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128. † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: litert-int4, where most use Q4_K_M.
Memory usage by engine
| Rank | Engine | Memory |
|---|---|---|
| 1 | Ollama†, approximate comparison, see details | 0.46 GiB |
| 2 | llama.cpp†, approximate comparison, see details | 0.46 GiB |
| 3 | uzu†, approximate comparison, see details | 0.48 GiB |
| 4 | MLX LM†, approximate comparison, see details | 0.50 GiB |
| 5 | Rapid-MLX†, approximate comparison, see details | 0.72 GiB |
| 6 | Magnitude†, approximate comparison, see details | 1.05 GiB |
| Not ranked | ExecuTorch | Not counted |
| Not ranked | LiteRT-LM | Not counted |
details
LFM2.5 350M on M5 Max · 64 GiB · 4-bit · Archive 256 input / 128 output · macOS 27
Memory ranks require comparable, calibrated measurements. Unavailable measurements remain unranked.
Ollama: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default.
llama.cpp: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default.
uzu: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mirai-m, where most use Q4_K_M.
MLX LM: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mlx-affine-4bit-g64, where most use Q4_K_M.
Rapid-MLX: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mlx-affine-4bit-g64, where most use Q4_K_M.
Magnitude: † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its phases were timed by the harness clock around the engine API, so they include transport that engine counters leave out.
ExecuTorch: Its output check failed: 6/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128. † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mlx-4w-g64, where most use Q4_K_M.
LiteRT-LM: Its output check failed: 6/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128. † Approximate comparison. The experiments use different effective KV cache precision: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: litert-int4, where most use Q4_K_M.
Every comparison
| Comparison | MLX LM | Rapid-MLX | uzu | Magnitude | Ollama | llama.cpp | LiteRT-LM | ZML | ExecuTorch |
|---|---|---|---|---|---|---|---|---|---|
| LFM2.5 350M on M5 Max · 64 GiB 4-bit · Archive 256 input / 128 output · macOS 27 | 100†, approximate | 100†, approximate | 90†, approximate | 82†, approximate | 67†, approximate | 62†, approximate | Not counted | Not run | Not counted |
| LFM2.5 350M on M5 Max · 64 GiB 8-bit · Archive 256 input / 128 output · macOS 27 | 100†, approximate | 98†, approximate | 97†, approximate | 93†, approximate | 77†, approximate | 75†, approximate | 24†, approximate | Not run | Not run |
| LFM2.5 350M on M5 Max · 64 GiB bf16 · Archive 256 input / 128 output · macOS 27 | 100†, approximate | 98†, approximate | Not run | 92†, approximate | 84†, approximate | 81†, approximate | Not run | 73†, approximate | Not run |
Each engine’s Speed score in each matched comparison, out of 100, or why it has none. How rankings work