AMD Ryzen AI Max+ PRO 495 with Radeon 8065S
160 GB decides what fits. 273.06 GB/s decides how fast it runs once it does.
Find one: Amazon ↗ · eBay (new and used) ↗Store links may pay us a commission. They never decide which card is suggested — the memory arithmetic does.
- Largest popular model that fitsMiniMax-M2.7229B params · Q4_K_M · needs 141.2 GB~25 tok/s About reading paceOpen in the calculator →
- Best fast pickQwen3.8-Flash-Next180B params · Q4_K_M · needs 111.6 GB~59 tok/s Faster than you readOpen in the calculator →
- Most downloaded that fitsQwen3.8-27B28B params · Q4_K_M · needs 18.6 GB~14 tok/s About reading paceOpen in the calculator →
At 8,192 tokens of context with the whole model in device memory. Speeds are estimates from memory bandwidth, not benchmarks run on this card.
258 of 320 fit entirely
258 of the 320 models the engine can size fit entirely in device memory at Q4_K_M where it is published, otherwise the nearest published format, and 8,192 tokens. Unified memory is one pool, so there is no second memory tier to spill into; the remaining 62 do not run at this context.
Featured models · Q4_K_M at 8,192 tokens, including offload
| Model | Needs | Verdict | Decode | Calculator |
|---|---|---|---|---|
| Qwen3.5-2B2.3B parameters | 2.32 GB | fits | ~172 tok/s | Open → |
| NVIDIA-Nemotron-3-Nano-4B-BF164.0B parameters | 3.46 GB | fits | ~80 tok/s | Open → |
| Qwen3-VL-8B-Instruct8.8B parameters | 7.04 GB | fits | ~36 tok/s | Open → |
| Qwen3.5-9B9.7B parameters | 7.05 GB | fits | ~44 tok/s | Open → |
| gemma-4-12B-it12B parameters | 9.54 GB | fits | ~26 tok/sAbout reading pace | Open → |
| gpt-oss-20b21B parameters | 13.8 GB | fits | ~85 tok/s | Open → |
| Qwen3.8-27B28B parameters | 18.6 GB | fits | ~14 tok/sAbout reading pace | Open → |
| Qwen3.6-27B28B parameters | 18.6 GB | fits | ~14 tok/sAbout reading pace | Open → |
| Kimi-Linear-48B-A3B-Instruct49B parameters | 30.4 GB | fits | ~114 tok/s | Open → |
| Qwen3.8-Flash-Next180B parameters | 111.6 GB | fits | ~59 tok/s | Open → |
| DeepSeek-V4-Flash-Vision-Exp305B parameters | 188.1 GB | short by 28.1 GB | does not run | Open → |
| GLM-5.3-Flash321B parameters | 198.3 GB | short by 38.3 GB | does not run | Open → |
These examples are selected from prominent labs using the catalogue’s latest Hugging Face 30-day downloads and repository-creation freshness signal, with newer releases guaranteed a place. Offloaded rows assume 32 GB of system RAM, and a speed is only shown for a row that runs. Fitting in memory is not the same as loading: whether the runtime and version you have supports each architecture and format on this machine has not been tested here. Each “Open” link carries the same model, format, context and RAM into the calculator.
Fits entirely in AMD Ryzen AI Max+ PRO 495 with Radeon 8065S memory — 258 of 320 sized models
The most demanding model that fits is Solar-Open2-250B at Q4_K_M: 152.1 GB of the 160 GB, leaving 7.87 GB spare.
Fully resident at 8,192 tokens (or the model’s own maximum, where that is shorter), offload off, at 273.06 GB/s. Each row is a run of the engine for this configuration; the rows start with current, prominent releases and “fits” is memory, not a tested runtime. Older or less prominent models remain available through this search and “Show all”.
Showing 40 of 258 models that fit.
| Model | Format | Needs | Spare | Decode | Downloads / 30d | Calculator |
|---|---|---|---|---|---|---|
| Qwen3.8-27BQwen · 28B params | Q4_K_M | 18.6 GB | 141.4 GB | ~14 tok/sAbout reading pace | 6.9M | Open → |
| Qwen3.8-Flash-NextNEWQwen · 180B params | Q4_K_M | 111.6 GB | 48.4 GB | ~59 tok/s | 1.4M | Open → |
| gemma-4-26B-A4B-itGoogle · 26B params | Q4_K_M | 16.8 GB | 143.2 GB | ~80 tok/s | 13M | Open → |
| gemma-4-31B-itGoogle · 31B params | Q4_K_M | 22.2 GB | 137.8 GB | ~10 tok/sAbout reading pace | 9.9M | Open → |
| Qwen3.5-9BQwen · 9.7B params | Q4_K_M | 7.05 GB | 153.0 GB | ~44 tok/s | 9M | Open → |
| Qwen3.5-4BQwen · 4.7B params | Q4_K_M | 3.98 GB | 156.0 GB | ~79 tok/s | 7.8M | Open → |
| Qwen3.6-35B-A3BQwen · 36B params | Q4_K_M | 23.1 GB | 136.9 GB | ~115 tok/s | 3.3M | Open → |
| gemma-4-12B-itGoogle · 12B params | Q4_K_M | 9.54 GB | 150.5 GB | ~26 tok/sAbout reading pace | 1.9M | Open → |
| Qwen3.6-27BQwen · 28B params | Q4_K_M | 18.6 GB | 141.4 GB | ~14 tok/sAbout reading pace | 2.5M | Open → |
| Qwen3.5-2BQwen · 2.3B params | Q4_K_M | 2.32 GB | 157.7 GB | ~172 tok/s | 4.9M | Open → |
| gemma-4-E4B-itGoogle · 8.0B params | Q4_K_M | 5.86 GB | 154.1 GB | ~70 tok/s | 4.4M | Open → |
| NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16NVIDIA · 32B params | Q4_K_M | 20.3 GB | 139.7 GB | ~11 tok/sAbout reading pace | 530K | Open → |
| NVIDIA-Nemotron-3-Nano-4B-BF16NVIDIA · 4.0B params | Q4_K_M | 3.46 GB | 156.5 GB | ~80 tok/s | 3.4M | Open → |
| gemma-4-E2B-itGoogle · 5.1B params | Q4_K_M | 4.02 GB | 156.0 GB | ~186 tok/s | 3M | Open → |
| Qwen3.5-0.8BQwen · 873M params | Q4_K_M | 1.46 GB | 158.5 GB | ~378 tok/s | 2.6M | Open → |
| Qwen3-VL-8B-InstructQwen · 8.8B params | Q4_K_M | 7.04 GB | 153.0 GB | ~36 tok/s | 14.6M | Open → |
| MiniMax-M2.7MiniMaxAI · 229B params | Q4_K_M | 141.2 GB | 18.8 GB | ~25 tok/sAbout reading pace | 1.1M | Open → |
| Qwen3.5-27BQwen · 28B params | Q4_K_M | 18.6 GB | 141.4 GB | ~14 tok/sAbout reading pace | 1.9M | Open → |
| North-Micro-Vision-InstructCohereLabs · 2.5B params | Q4_K_M | 2.99 GB | 157.0 GB | ~113 tok/s | 180.7K | Open → |
| Qwen3.5-35B-A3BQwen · 36B params | Q4_K_M | 23.1 GB | 136.9 GB | ~115 tok/s | 1.6M | Open → |
| NVIDIA-Nemotron-3-Super-120B-A12B-BF16NVIDIA · 124B params | Q4_K_M | 76.9 GB | 83.1 GB | ~2.8 tok/sVery slow | 1.2M | Open → |
| granite-4.2-8bIBM · 8.8B params | Q4_K_M | 7.49 GB | 152.5 GB | ~33 tok/s | 128.4K | Open → |
| GLM-4.7-Flashzai-org · 31B params | Q4_K_M | 20.4 GB | 139.6 GB | ~11 tok/sAbout reading pace | 1.8M | Open → |
| granite-4.1-3bIBM · 3.4B params | Q4_K_M | 3.72 GB | 156.3 GB | ~77 tok/s | 521.8K | Open → |
| LFM2.5-2.6BLiquidAI · 2.7B params | Q4_K_M | 2.61 GB | 157.4 GB | ~123 tok/s | 108.1K | Open → |
| Qwen3-VL-4B-InstructQwen · 4.4B params | Q4_K_M | 4.72 GB | 155.3 GB | ~57 tok/s | 3.5M | Open → |
| granite-4.1-30bIBM · 29B params | Q4_K_M | 20.4 GB | 139.6 GB | ~11 tok/sAbout reading pace | 301.5K | Open → |
| Qwen3.5-122B-A10BQwen · 125B params | Q4_K_M | 77.9 GB | 82.1 GB | ~42 tok/s | 512.6K | Open → |
| Qwen3-VL-2B-InstructQwen · 2.1B params | Q4_K_M | 3.05 GB | 157.0 GB | ~104 tok/s | 2.8M | Open → |
| granite-4.2-3bIBM · 3.7B params | Q4_K_M | 3.72 GB | 156.3 GB | ~77 tok/s | 50.6K | Open → |
| LFM2.5-230MLiquidAI · 230M params | Q4_K_M | 1.05 GB | 158.9 GB | ~859 tok/s | 88.6K | Open → |
| Qwen3-0.6BQwen · 752M params | Q4_K_M | 2.22 GB | 157.8 GB | ~160 tok/s | 29.7M | Open → |
| Qwen3-Coder-NextQwen · 80B params | Q4_K_M | 50.0 GB | 110.0 GB | ~76 tok/s | 596.3K | Open → |
| gpt-oss-20bOpenAI · 21B params | Q4_K_M | 13.8 GB | 146.2 GB | ~85 tok/s | 6.6M | Open → |
| MiniMax-M2.5MiniMaxAI · 229B params | Q4_K_M | 141.2 GB | 18.8 GB | ~25 tok/sAbout reading pace | 444.8K | Open → |
| granite-4.2-30bIBM · 29B params | Q4_K_M | 20.7 GB | 139.3 GB | ~11 tok/sAbout reading pace | 35.8K | Open → |
| granite-4.1-8bIBM · 8.8B params | Q4_K_M | 7.49 GB | 152.5 GB | ~33 tok/s | 179.3K | Open → |
| NVIDIA-Nemotron-3-Nano-30B-A3B-BF16NVIDIA · 32B params | Q4_K_M | 20.3 GB | 139.7 GB | ~11 tok/sAbout reading pace | 875.2K | Open → |
| gpt-oss-120bOpenAI · 117B params | Q4_K_M | 71.9 GB | 88.1 GB | ~60 tok/s | 4.5M | Open → |
| LFM2.5-VL-3BLiquidAI · 3.1B params | Q4_K_M | 2.85 GB | 157.1 GB | ~123 tok/s | 26.7K | Open → |
Sized at the format most people actually download, not at FP16. The catalogue ordered by downloads →
Find the AMD Ryzen AI Max+ PRO 495 with Radeon 8065S: Amazon ↗ · eBay (new and used) ↗Store links may pay us a commission. They never decide which card is suggested — the memory arithmetic does.
The specification behind every figure
What the manufacturer publishes for this device, and the pages it was read from.
Manufacturer specification
| Memory scope | shared-system |
|---|---|
| Capacity | 160 GB |
| Published options | 160 GB |
| Bandwidth | 273.06 GB/s |
| Memory type | LPDDR5x system memory |
| Bus width | 256-bit |
| FP32 peak | Not published |
| Dense matrix peak | Not published without sparsity |
| cTDP, configurable 45-120 W | 120 W |
Source ledger
Caveats
- The capacity catalogued is graphics memory, not system memory. AMD's specification page gives "Max. Memory 192 GB" for the processor; its blog footnote GRHP-01 gives "up to 160 GB dedicated graphics memory". A model has to fit in what the graphics processor can address, so 160 GB is what verdicts are taken against and the 192 GB system total is recorded here rather than counted.
- AMD publishes no memory bandwidth for this part. 273.06 GB/s is the arithmetic on the two figures it does publish — a 256-bit LPDDR5x interface at LPDDR5x-8533 — and is labelled derived for that reason.
- A configurable-TDP mobile and small-form-factor part: 120 W is the top of AMD's published 45-120 W range, and the machine a reader buys may be configured lower, which changes sustained throughput.
- Announced for third-quarter 2026 availability through OEM partners; this catalogue has not verified a shipping configuration from a maker, so the memory options are AMD's processor maximum rather than a machine anyone sells.
- The sibling entry amd-ryzen-ai-max-plus-395-strix-halo records its 128 GB as maximum system memory with a caveat, because AMD publishes no separate graphics-memory ceiling for that part. The two entries therefore count different things, which is why each says which.
- No dense matrix throughput is published for this device in a form this catalogue accepts, so no compute roof is priced and time to first token is withheld.
Published capacity is a hardware ceiling, not guaranteed free runtime memory. The calculator shows the runtime reserve separately rather than folding it into a single number.
Run the diagnostic on the AMD Ryzen AI Max+ PRO 495 with Radeon 8065S →