Comparison Counted
Which engine runs LFM2.5 2.6B best on M5 Max · 64 GiB?
- Device
- M5 Max · 64 GiB (Apple silicon Macs)
- OS
- macOS 27
- Model
- LFM2.5 2.6B, 2.6B parameters
- Weights
- 8-bit
- Workload
- Archive 256 input / 128 output: 256 tokens in, 128 out
- Decoding
- Greedy, batch 1, context 4,096
- Power
- Plugged in
Ranking
Balanced: Decode speed, time to first token and peak memory, weighted equally.
| # | Engine | Balanced score out of 100 |
|---|---|---|
| 1 | MLX LM†, an approximate comparison, see details | 99 /100 Balanced score |
| 2 | Ollama†, an approximate comparison, see details | 98 /100 Balanced score |
| 3 | Rapid-MLX†, an approximate comparison, see details | 92 /100 Balanced score |
| 4 | llama.cpp†, an approximate comparison, see details | 91 /100 Balanced score |
| Not ranked in this view | ||
| Not ranked | Magnitude | Unsupported Unsupported in 1 comparison. 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). |
| Not ranked | uzu | Unsupported Unsupported in 1 comparison. uzu can't run the frozen prompts: Error: uzu counted 23 input tokens, not the frozen prompt's 26 (measured in qualification). |
| Not ranked | LiteRT-LM | Not counted 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. |
details
Balanced Decode speed, time to first token and peak memory, weighted equally.
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: litert-lm default, f16, model-dtype. The experiments here use different 8-bit schemes: litert-int8, Q8_0, mlx-affine-8bit-g64.
† Ollama: approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 8-bit schemes: litert-int8, Q8_0, mlx-affine-8bit-g64.
† Rapid-MLX: approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 8-bit schemes: litert-int8, Q8_0, mlx-affine-8bit-g64.
† llama.cpp: approximate comparison. The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 8-bit schemes: litert-int8, Q8_0, mlx-affine-8bit-g64.
Every experiment
Medians over valid trials, with the lowest and highest beneath. Eligible experiments are complete and passed their output check. The ranking above selects each engine's configuration for this objective when its required metrics are measured.
| Experiment | Ranking eligibility | Decode tok/s, higher is better | First token ms, lower is better | Memory GiB, lower is better | Prefill tok/s, higher is better | Valid trials |
|---|---|---|---|---|---|---|
| MLX LM 0.32.2 · Metal · mlx-affine-8bit-g64 | Eligible, approximately: The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 8-bit schemes: litert-int8, Q8_0, mlx-affine-8bit-g64. | 163.1 range158.1–163.9 | 53.2 range53.0–55.1 | 3.02 range2.99–3.07 | 4,868 range4,703–4,882 | 6of 6 planned |
| Ollama 0.40.13 · Metal · Q8_0 | Eligible, approximately: The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 8-bit schemes: litert-int8, Q8_0, mlx-affine-8bit-g64. | 155.1 range155.0–155.2 | 52.7 range52.1–52.9 | 2.96 range2.96–2.97 | 4,995 range4,957–5,051 | 6of 6 planned |
| Rapid-MLX 1.0.2 · Metal · mlx-affine-8bit-g64 | Eligible, approximately: The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 8-bit schemes: litert-int8, Q8_0, mlx-affine-8bit-g64. | 162.6 range161.1–163.6 | 61.2 range58.9–61.9 | 3.28 range3.23–3.41 | 4,220 range4,175–4,389 | 6of 6 planned |
| llama.cpp b12110 · Metal · Q8_0 | Eligible, approximately: The experiments use different effective KV cache precision: litert-lm default, f16, model-dtype. The experiments here use different 8-bit schemes: litert-int8, Q8_0, mlx-affine-8bit-g64. | 148.8 range116.0–150.1 | 62.8 range61.2–63.7 | 2.96 range2.96–2.98 | 4,126 range4,062–4,241 | 6of 6 planned |
| LiteRT-LM 0.19.1 · CPU · litert-int8 | Complete 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. | 60.3 range60.0–60.5 | 541 range539–542 | 3.42 range3.42–3.42 | 515.9 range514.4–517.4 | 6of 6 planned |
| Magnitude 0.2.17 · Metal · Q8_0 | Unsupported 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). | – | – | – | – | 0of 6 planned |
| uzu 0.6.17 · Metal · mirai-l | Unsupported uzu can't run the frozen prompts: Error: uzu counted 23 input tokens, not the frozen prompt's 26 (measured in qualification). | – | – | – | – | 0of 6 planned |
Every trial, metric by metric
The line runs from an experiment's lowest to its highest valid trial, the dot is the median and each tick is one trial. Every scale starts at zero.
Caveats
- Publisher cards identify the base model but do not always pin the exact export source checkpoint revision.
- Engines may differ in quantization scheme, effective KV cache precision, kernels, and activations.
About this comparison
Worked out from its experiments: every experiment here shares the device, the OS major version, the power state, the base model, the quantization class and the workload. Measured under methodology v0.6. Comparison id glenn-m5-max-64gb--macos-27--lfm2.5-2.6b--8-bit--archive-256-128.