AMD · unified x86

AMD Ryzen AI Max+ 395 (Radeon 8060S)

AMD Ryzen AI Max+ 395 (Radeon 8060S) has 64 GB of VRAM at 256 GB/s — about 44.64 GiB usable after driver and compositor overhead. 2029 of 2118 indexed models fit at 128K context with q4_0 KV. Note only 48 GB of its 64 GB is allocatable to the GPU.

Spec sheet· bandwidth, theoreticalFrom the file· fit from summed bytesPredicted· speed
Memory
64 GB
LPDDR5X-8000
Bandwidth
256 GB/s
256-bit bus
Tensor FP16
—
dense
TDP
120 W
Context4K8K16K32K64K128K
KV cachef16q8_0q4_0quantizing the KV cache is a ~2× lever on the dominant term at long context
text 1741vision language 184image 2audio tts 21video 16audio asr 39embedding 26

What fits at 128K context

largest quantization that fits, per model · 2029 of 2118 indexed
ModelBest quantParamsWeights●KV●Total in memory◐Headroom◐tok/s◐
Qwen3.5-88BMoEI1-Q4_087.7B46.41 GiB0.84 GiB47.99 GiB0.01 GiB14±37%
Devstral-2-123B-Instruct-2512IQ2_XS125B34.75 GiB12.38 GiB47.98 GiB0.02 GiB3±25%
Mistral-Medium-3.5-128BI1-IQ2_XS128B34.75 GiB12.38 GiB47.98 GiB0.02 GiB3±25%
XORTRON-NXTXPRTXXLI1-IQ2_XS128B34.75 GiB12.38 GiB47.98 GiB0.02 GiB3±25%
Mistral-Small-4-119B-2603MoEIQ3_XXS119B46.44 GiB0.79 GiB47.97 GiB0.03 GiB16±37%
Hunyuan-A13B-InstructMoEIQ4_NL80.4B42.77 GiB4.50 GiB47.96 GiB0.04 GiB3±25%
GPT-NeoX-20B-ErebusI1-Q3_K_M20.6B10.03 GiB37.13 GiB47.95 GiB0.05 GiB3±25%
GLM-4.5VMoEI1-Q2_K_S108B40.74 GiB6.47 GiB47.94 GiB0.06 GiB7±37%
Step-3.5-Flash-REAP-121B-A11BI1-IQ2_S121B33.40 GiB13.79 GiB47.91 GiB0.09 GiB3±25%
deepseek-llm-67b-chatI1-IQ4_XS67.4B33.71 GiB13.36 GiB47.87 GiB0.13 GiB3±25%
deepseek-llm-67b-baseI1-IQ4_XS67.4B33.71 GiB13.36 GiB47.87 GiB0.13 GiB3±25%
openbuddy-deepseek-67b-v15.3-4kI1-IQ4_XS67.4B33.71 GiB13.36 GiB47.87 GiB0.13 GiB3±25%
command-a-plus-05-2026-bf16MoEIQ1_S219B45.87 GiB1.24 GiB47.82 GiB0.18 GiB12±37%
HunyuanImage-2.1Q5_017.5B47.04 GiB0.00 GiB47.79 GiB0.21 GiB3±25%
Qwen3-Coder-NextMoEQ4_K_S79.7B43.71 GiB3.38 GiB47.77 GiB0.23 GiB12±37%
Qwen3-Next-80B-A3B-ThinkingMoEQ4_K_S81.3B43.71 GiB3.38 GiB47.77 GiB0.23 GiB12±37%
Qwen3-Next-80B-A3B-InstructMoEQ4_K_S81.3B43.71 GiB3.38 GiB47.77 GiB0.23 GiB12±37%
Qwen3-Coder-Next-REAMMoEI1-Q6_K60.3B46.22 GiB0.84 GiB47.75 GiB0.25 GiB17±37%
c4ai-command-r-plus-08-2024IQ3_XXS104B37.87 GiB9.00 GiB47.75 GiB0.25 GiB3±25%
Maenad-70BIQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
DeepSeek-R1-Distill-Llama-70B-Uncensored-v2-Unbiased-ReasonerIQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
Strawberrylemonade-L3-70B-v1.2IQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
Llama-3.3_70_b_uncensored_continuedIQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
Llama-3.3-70B-Instruct-abliteratedIQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
grok-oss-Revenant-70BIQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
Hermes-4-70B-hereticIQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
Golem-70B-v1bIQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
DeepSeek-R1-Distill-Llama-70B-hereticIQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
Infinity-Instruct-7M-Gen-Llama3_1-70BIQ4_XS70.6B35.64 GiB11.25 GiB47.72 GiB0.28 GiB3±25%
ALIA-40b-fc-2606Q8_040.4B40.02 GiB6.75 GiB47.58 GiB0.42 GiB3±25%
ALIA-40b-instruct-2606Q8_040.4B40.02 GiB6.75 GiB47.58 GiB0.42 GiB3±25%
Laguna-S-2.1MoEUD-IQ3_S118B45.10 GiB1.73 GiB47.55 GiB0.45 GiB13±37%
Llama-4-Scout-17B-16E-InstructMoEKV unresolvedQ2_K109B40.03 GiB6.75 GiB47.51 GiB0.49 GiB7±37%
Apertus-70B-Instruct-2509IQ4_XS70.6B35.36 GiB11.25 GiB47.49 GiB0.51 GiB3±25%
MiniMax-M2.1-REAP-139B-A10BMoEI1-IQ2_XS139B38.06 GiB8.72 GiB47.46 GiB0.54 GiB6±37%
m51Lab-MiniMax-M2.7-REAP-139B-A10BMoEI1-IQ2_XS139B38.06 GiB8.72 GiB47.46 GiB0.54 GiB6±37%
Delphi-25B-SimpleRL-MathI1-IQ3_XXS25.0B9.03 GiB37.65 GiB47.46 GiB0.54 GiB3±25%
GLM-4.6VMoEIQ2_M108B40.26 GiB6.47 GiB47.45 GiB0.55 GiB7±37%
Hermes-4-70BIQ4_XS70.6B35.33 GiB11.25 GiB47.41 GiB0.59 GiB3±25%
Llama-3.3-70B-InstructIQ4_XS70.6B35.33 GiB11.25 GiB47.41 GiB0.59 GiB3±25%
DeepSeek-R1-Distill-Llama-70BIQ4_XS70.6B35.33 GiB11.25 GiB47.41 GiB0.59 GiB3±25%
Qwen3-42B-A3B-2507-Thinking-Abliterated-uncensored-TOTAL-RECALL-v2-Medium-MASTER-CODERMoEQ8_042.4B41.98 GiB4.71 GiB47.39 GiB0.61 GiB8±37%
calme-2.4-llama3-70bIQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
calme-2.2-llama3-70bIQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
Rombos-LLM-70b-Llama-3.3I1-IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
L3.3-Electra-R1-70bI1-IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
L3.3-70B-Magnum-v4-SEIQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
Latxa-Llama-3.1-70B-Instruct-v2I1-IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
Llama-3.1-Nemotron-70B-Instruct-HFI1-IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
L3.3-70B-Euryale-v2.3I1-IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
Hermes-3-Llama-3.1-70BIQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
Llama-3.1-70BIQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
Anubis-70B-v1.2IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
DeepSeek-R1-Distill-Llama-70B-abliteratedI1-IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
llama-3-firefunction-v2IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
Legion-V2.1-LLaMa-70BI1-IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
Assistant_Pepe_70BI1-IQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
Tess-R1-Limerick-Llama-3.1-70BIQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
SEMIKONG-70BIQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
functionary-medium-v3.2KV unresolvedIQ4_XS70.6B35.30 GiB11.25 GiB47.38 GiB0.62 GiB3±25%
●From the file◐Predictedwhat these mean

Speed is modeled, not measured: decode is memory-bandwidth bound, so tokens per second is bytes read per token against achievable bandwidth. Mixture-of-experts models carry a wider band because only the routed experts are read each step, and few have been measured publicly.

Questions people ask

What AI models can a AMD Ryzen AI Max+ 395 (Radeon 8060S) run?
2029 of 2118 indexed open-weight models fit a AMD Ryzen AI Max+ 395 (Radeon 8060S) at 131,072 context with q4_0 KV cache, the largest being Qwen3.5-88B at I1-Q4_0. That covers text, vision-language, image, video and speech models.
How much usable memory does a AMD Ryzen AI Max+ 395 (Radeon 8060S) actually have?
Its nameplate is 64 GB, but about 44.64 GiB is available to a model once driver and compositor overhead is accounted for, and only 48 GB of the pool can be allocated to the GPU at all.
Is a AMD Ryzen AI Max+ 395 (Radeon 8060S) fast for local AI?
Its memory bandwidth is 256 GB/s, and that figure — not teraflops — is what governs token generation speed. Capacity decides what you can run; bandwidth decides how fast it runs.