Kimi K2.7-Code

See Kimi K2.7-Code in action: demonstration videos

Tested on an M3 Ultra 512 GiB RAM using Inferencer app

• Text inference: ~21.5 tokens/s @ 4000 tokens ~426.5 GiB
• Vision inference: ~22.6 tokens/s @ 4000 tokens ~426.8 GiB

👁 Screenshot 1
👁 Screenshot 2
👁 Screenshot 3

Q3.5-INF uses the data-agnostic INF method tuned to yield maximum general accuracy within a 512 GiB memory budget.

Due to system memory and time constraints, the figures shown below are from the conversion of Kimi K2.6.

Quantization (bpw)	Perplexity	Token Accuracy	Missed Divergence
Q3.5	1.1328125	94.92%	42.71%
Q3.5-INF	1.078125	96.67%	22.04%
Q3.6	1.1484375	94.72%	48.72%
Base	Untested	100%	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 our demonstration videos or visit Kimi-K2.7-Code.

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.

Base model

moonshotai/Kimi-K2.7-Code

(15)