Explore benchmarks
Filter by model, hardware, workload and quantization. Every score links to the comparisons and experiments behind it.
Top Engines
Not enough eligible results to rank.
Speed
| Rank | Engine | Prefill |
|---|---|---|
| Not ranked | LiteRT-LM†, approximate comparison, see details | 166.0 tok/s |
| Not ranked | llama.cpp | Not counted |
| Not ranked | Magnitude | Unsupported |
| Not ranked | MLX LM | Not counted |
| Not ranked | Ollama | Not counted |
| Not ranked | Rapid-MLX | Not counted |
| Not ranked | uzu | Unsupported |
details
LFM2.5 2.6B 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.
LiteRT-LM: † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
llama.cpp: Its output check failed: 8/9 of the checks the model passes at BF16; needs 8 and the mandatory ones (needle-start, continuation-128). Left out, failed by every BF16 configuration: color, sort, needle-end. † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
Magnitude: Magnitude's chat API renders LFM2.5 2.6B's own template, which always opens a thinking block, so it streams reasoning and can't run the closed-think prompts (measured in qualification).
MLX LM: Its output check failed: 7/9 of the checks the model passes at BF16; needs 8 and the mandatory ones (needle-start, continuation-128). Left out, failed by every BF16 configuration: color, sort, needle-end. † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
Ollama: Its output check failed: 8/9 of the checks the model passes at BF16; needs 8 and the mandatory ones (needle-start, continuation-128). Left out, failed by every BF16 configuration: color, sort, needle-end. † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
Rapid-MLX: Its output check failed: 7/9 of the checks the model passes at BF16; needs 8 and the mandatory ones (needle-start, continuation-128). Left out, failed by every BF16 configuration: color, sort, needle-end. † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
uzu: uzu can't run the frozen prompts: Error: uzu counted 23 input tokens, not the frozen prompt's 26 (measured in qualification).
Memory usage by engine
| Rank | Engine | Memory |
|---|---|---|
| Not ranked | LiteRT-LM†, approximate comparison, see details | 2.24 GiB |
| Not ranked | llama.cpp | Not counted |
| Not ranked | Magnitude | Unsupported |
| Not ranked | MLX LM | Not counted |
| Not ranked | Ollama | Not counted |
| Not ranked | Rapid-MLX | Not counted |
| Not ranked | uzu | Unsupported |
details
LFM2.5 2.6B on M5 Max · 64 GiB · 4-bit · Archive 256 input / 128 output · macOS 27
Memory ranks require comparable, calibrated measurements. Unavailable measurements remain unranked.
LiteRT-LM: † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
llama.cpp: Its output check failed: 8/9 of the checks the model passes at BF16; needs 8 and the mandatory ones (needle-start, continuation-128). Left out, failed by every BF16 configuration: color, sort, needle-end. † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
Magnitude: Magnitude's chat API renders LFM2.5 2.6B's own template, which always opens a thinking block, so it streams reasoning and can't run the closed-think prompts (measured in qualification).
MLX LM: Its output check failed: 7/9 of the checks the model passes at BF16; needs 8 and the mandatory ones (needle-start, continuation-128). Left out, failed by every BF16 configuration: color, sort, needle-end. † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
Ollama: Its output check failed: 8/9 of the checks the model passes at BF16; needs 8 and the mandatory ones (needle-start, continuation-128). Left out, failed by every BF16 configuration: color, sort, needle-end. † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
Rapid-MLX: Its output check failed: 7/9 of the checks the model passes at BF16; needs 8 and the mandatory ones (needle-start, continuation-128). Left out, failed by every BF16 configuration: color, sort, needle-end. † Approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 4-bit schemes: litert-int4, Q4_K_M, mlx-affine-4bit-g64.
uzu: uzu can't run the frozen prompts: Error: uzu counted 23 input tokens, not the frozen prompt's 26 (measured in qualification).
Every comparison
| Comparison | LiteRT-LM | llama.cpp | Magnitude | MLX LM | Ollama | Rapid-MLX | uzu |
|---|---|---|---|---|---|---|---|
| LFM2.5 2.6B on M5 Max · 64 GiB 4-bit · Archive 2048 input / 128 output · macOS 27 Not counted: fewer than two scored engines | Measured†, approximate | Not counted | Unsupported | Not counted | Not counted | Not counted | Unsupported |
| LFM2.5 2.6B on M5 Max · 64 GiB 4-bit · Archive 256 input / 128 output · macOS 27 Not counted: fewer than two scored engines | Measured†, approximate | Not counted | Unsupported | Not counted | Not counted | Not counted | Unsupported |
Each engine’s Speed score in each matched comparison, out of 100, or why it has none. How rankings work