Home › Models › Inkling Small › AMD Ryzen AI Max+ 395 (Radeon 8060S)
Inkling Small on the AMD Ryzen AI Max+ 395 (Radeon 8060S)
Inkling Small runs at about 21.0 tokens/s on the AMD Ryzen AI Max+ 395 (Radeon 8060S) (measured, UD-IQ2_M, ROCm, llama-bench).
UD-IQ2_M · ROCm
same run
For comparison, our size-scaling estimate for this combination is 28.0 tok/s.
Every run
| Quant | Backend | Generate tok/s | Prompt tok/s | Test | Build / by / notes | Source |
|---|---|---|---|---|---|---|
| UD-IQ2_M | ROCm | 21.0 | 122 | tg32 | 1e6f9e4a5 128 GB Strix Halo, ROCm 7.14, -ub 256, model fully in memory; llama.cpp from upstream PR #25731 (unmerged at the time), built from 1e6f9e4a5. | llama.cpp PR #25731 comment by MatioLK (GitHub) |
tg128 = generating 128 tokens; pp512 = reading a 512-token prompt (llama-bench). LocalScore averages several prompt and output lengths, so its numbers run lower than tg128. "decode, llama-server" = single-stream generation speed measured against a running llama-server (no speculative decoding).
No card yet?
Try it on a rented GPU (RunPod)Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.
All results for the AMD Ryzen AI Max+ 395 (Radeon 8060S) · Inkling Small on other hardware
Embed this result
HTML
<a href="https://rigbench.pages.dev/model/inkling-small/on/amd-ryzen-ai-maxplus-395-radeon-8060s/"><img src="https://rigbench.pages.dev/badge/inkling-small--amd-ryzen-ai-maxplus-395-radeon-8060s.svg" alt="Inkling Small · AMD Ryzen AI Max+ 395 (Radeon 8060S): 21.0 tok/s (UD-IQ2_M, measured)"></a>
Markdown (GitHub README, forums)
[](https://rigbench.pages.dev/model/inkling-small/on/amd-ryzen-ai-maxplus-395-radeon-8060s/)