readygpu

Kimi Linear 48B-A3B

49.1B total, 3B active · 1024k context · mit · released 2025-10-30

chatcoding

What to run it on in the UK

Fully in memory at 8k context, priced at retailers today.

Something we price holds it fully in memory, but none reaches 15 tok/s, our bar for recommending it. The table below has the speeds.

All hardware that runs it

Everything we price, best value first. Two cards split the model by layers.

HardwareQuantFitsDecodePricetok/s per 1,000
NVIDIA GeForce RTX 5090Q4_K_Mfits343 tok/s––
Mac Studio M4 Max 128GBQ8_0fits65 tok/s––
Mac mini M4 Pro 64GBQ6_Kfits41 tok/s––
Framework Desktop Ryzen AI Max+ 395 128GBQ8_0fits26 tok/s––
Mac Studio M3 Ultra 512GBQ8_0fits97 tok/s––
NVIDIA DGX Spark 128GBQ8_0fits31 tok/s––
GMKtec EVO-X2 Ryzen AI Max+ 395 128GBQ8_0fits26 tok/s––
Mac Studio M3 Ultra 256GBQ8_0fits97 tok/s––
Mac mini M4 Pro 48GBQ5_K_Mfits47 tok/s––
Mac Studio M4 Max 64GBQ6_Kfits83 tok/s––
Mac Studio M3 Ultra 96GBQ8_0fits97 tok/s––
NVIDIA RTX PRO 6000 BlackwellQ8_0fits202 tok/s––
AMD Radeon AI PRO R9700Q4_K_Mfits109 tok/s––
ASUS Ascent GX10 128GBQ8_0fits31 tok/s––
Framework Desktop Ryzen AI Max+ 395 64GBQ6_Kfits33 tok/s––
NVIDIA GeForce RTX 3090Q4_K_Moffload62 tok/s––
NVIDIA GeForce RTX 4090Q4_K_Moffload63 tok/s––
NVIDIA GeForce RTX 5070 TiQ4_K_Moffload31 tok/s––
NVIDIA GeForce RTX 5060 Ti 16GBQ4_K_Moffload29 tok/s––
AMD Radeon RX 7900 XTXQ4_K_Moffload58 tok/s––
NVIDIA GeForce RTX 3060 12GBQ4_K_Moffload23 tok/s––
AMD Radeon RX 9070 XTQ4_K_Moffload29 tok/s––
NVIDIA GeForce RTX 4060 Ti 16GBQ4_K_Moffload26 tok/s––
NVIDIA GeForce RTX 5080Q4_K_Moffload32 tok/s––
NVIDIA GeForce RTX 4070 Ti SuperQ4_K_Moffload30 tok/s––
NVIDIA GeForce RTX 5070Q4_K_Moffload25 tok/s––
AMD Radeon RX 9060 XT 16GBQ4_K_Moffload26 tok/s––
NVIDIA GeForce RTX 4070 SuperQ4_K_Moffload24 tok/s––
NVIDIA GeForce RTX 4080 SuperQ4_K_Moffload31 tok/s––
AMD Radeon RX 7800 XTQ4_K_Moffload29 tok/s––
AMD Radeon RX 7900 XTQ4_K_Moffload39 tok/s––
Intel Arc B580Q4_K_Moffload21 tok/s––
NVIDIA GeForce RTX 3090 TiQ4_K_Moffload63 tok/s––
AMD Radeon RX 7600 XTQ4_K_Moffload25 tok/s––
NVIDIA GeForce RTX 3080 10GBQ4_K_Moffload23 tok/s––
Intel Arc Pro B60Q4_K_Moffload39 tok/s––
NVIDIA Tesla P40Q4_K_Moffload42 tok/s––
Intel Arc A770 16GBQ4_K_Moffload26 tok/s––

2 other cards and machines we price cannot hold it at this context.

Memory needed

Weights, KV cache and runtime buffers at each context length (KV cache at fp16).

QuantFile@8k@32k@64k@128k
Q4_K_Mgguf28.0 GB28.8 GB29.2 GB29.6 GB30.5 GB
mlx-4bitmlx25.7 GB26.6 GB26.9 GB27.3 GB28.2 GB
Q5_K_Mgguf32.7 GB33.5 GB33.9 GB34.3 GB35.2 GB
Q6_Kgguf37.7 GB38.5 GB38.9 GB39.3 GB40.2 GB
mlx-8bitmlx48.6 GB49.5 GB49.8 GB50.2 GB51.1 GB
Q8_0gguf48.7 GB49.5 GB49.8 GB50.3 GB51.1 GB

Run it

Copy a command for the runtime you use.

  • Q4_K_M
    • Ollamaollama run hf.co/bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF:Q4_K_M
    • LM Studiolms get bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF@q4_k_m
    • llama.cppllama-server -hf bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF:Q4_K_M
  • mlx-4bit
    • LM Studiolms get mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit
    • MLXmlx_lm.server --model mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit
  • Q5_K_M
    • Ollamaollama run hf.co/bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF:Q5_K_M
    • LM Studiolms get bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF@q5_k_m
    • llama.cppllama-server -hf bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF:Q5_K_M
  • Q6_K
    • Ollamaollama run hf.co/bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF:Q6_K
    • LM Studiolms get bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF@q6_k
    • llama.cppllama-server -hf bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF:Q6_K
  • mlx-8bit
    • LM Studiolms get mlx-community/Kimi-Linear-48B-A3B-Instruct-8bit
    • MLXmlx_lm.server --model mlx-community/Kimi-Linear-48B-A3B-Instruct-8bit
  • Q8_0
    • Ollamaollama run hf.co/bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF:Q8_0
    • LM Studiolms get bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF@q8_0
    • llama.cppllama-server -hf bartowski/moonshotai_Kimi-Linear-48B-A3B-Instruct-GGUF:Q8_0