inferencerlabs/Kimi-K2.5-MLX-4.2bit

See Kimi-K2.5 MLX in action - demonstration video

Tested on a M3 Ultra 512GB + M4 Max 128GB using distributed compute over Wifi with Inferencer app v1.9.5

• Single inference ~22.2 tokens/s @ 1000 tokens
• Memory usage: ~513 GiB
• For a larger context window you can expand the RAM limit:
  • sudo sysctl iogpu.wired_limit_mb=510000

q4.2bit quant typically achieves 1.39062 perplexity in our coding test

Quantization	Perplexity	Token Accuracy	Missed Divergence
q3.5	168.0	43.45%	72.57%
q3.6	1.59375	83.75%	41.11%
q3.8	1.49218	86.65%	36.79%
q4.2	1.39062	89.30%	31.70%
q4.5	1.33593	91.65%	27.61%
q6.5	1.21875	96.95%	12.03%
Base	1.20312	100.0%	0.000%

• Perplexity: Measures the confidence for predicting base tokens (lower is better)
• Token Accuracy: The percentage of correctly generated base tokens
• Missed Divergence: Measures severity of misses; how much the token was missed by

Quantized with a modified version of MLX

For more details see demonstration video or visit Kimi-K2.5.

Disclaimer

We are not the creator, originator, or owner of any model listed. Each model is created and provided by third parties. Models may not always be accurate or contextually appropriate. You are responsible for verifying the information before making important decisions. We are not liable for any damages, losses, or issues arising from its use, including data loss or inaccuracies in AI-generated content.