What Models?

Pick the right model for your GPU — in seconds.

What GPU do you need to run LFM2.5 2.6B?

2.7B params 128K max context Tool use

The smallest GPU memory that runs LFM2.5 2.6B well is 2 GB (Q4_K_M, e.g. T400). 132 of 132 GPUs run it well.

VRAM needed per quantization

QuantizationWeights16K context32K contextFull 128K
Q4_K_M1.6 GB1.8 GB2.0 GB3.5 GB
Q8_02.7 GB2.9 GB3.2 GB4.6 GB

Weights plus fp16 KV cache. Leave ~1 GB headroom for the inference engine.

Cheapest GPUs that can run LFM2.5 2.6B

Lowest launch MSRP among GPUs that run its smallest quantization (Q4_K_M) well, with at least 16K context.

As an Amazon Associate, I earn from qualifying purchases.

GPUs that can run LFM2.5 2.6B

Best quantization that fits on each GPU, with the context and speed it gets. For chips with several memory sizes, the smallest that works is shown.

NVIDIA

AMD

Intel

Apple

  • M1 8 GB Q8_0 128K context · ~18 tok/s
  • M2 8 GB Q8_0 128K context · ~27 tok/s
  • M3 8 GB Q8_0 128K context · ~27 tok/s
  • M1 Pro 16 GB Q8_0 128K context · ~54 tok/s
  • M2 Pro 16 GB Q8_0 128K context · ~54 tok/s
  • M4 16 GB Q8_0 128K context · ~33 tok/s
  • M5 16 GB Q8_0 128K context · ~42 tok/s
  • M6 16 GB Q8_0 128K context · ~42 tok/s
  • M3 Pro 18 GB Q8_0 128K context · ~41 tok/s
  • M4 Pro 24 GB Q8_0 128K context · ~74 tok/s
  • M5 Pro 24 GB Q8_0 128K context · ~83 tok/s
  • M1 Max 32 GB Q8_0 128K context · ~109 tok/s
  • M2 Max 32 GB Q8_0 128K context · ~109 tok/s
  • M3 Max 36 GB Q8_0 128K context · ~82 tok/s
  • M4 Max 36 GB Q8_0 128K context · ~111 tok/s
  • M5 Max 36 GB Q8_0 128K context · ~125 tok/s
  • M1 Ultra 64 GB Q8_0 128K context · ~217 tok/s
  • M2 Ultra 64 GB Q8_0 128K context · ~217 tok/s
  • M3 Ultra 96 GB Q8_0 128K context · ~223 tok/s
  • M5 Ultra 96 GB Q8_0 128K context · ~326 tok/s
Check your exact setup in the full tool →