AMD · Gorgon Halo · 192 GB · Qwen3.8 27B · Q4_K_M: ~8.0 tok/s
AMD Ryzen AI Max+ PRO 495 desktop, 192 GB
The Strix Halo design with 192 GB instead of 128, and up to 160 GB of it for the GPU: enough for a 235B mixture-of-experts model at Q4. Bandwidth barely moves, so it holds bigger models at the same speed. Sold by Framework, Minisforum and AMD itself.
Lowest price right now
$8,099.99
eBay · free shipping · $42 per GB · observed 2026-10-18 02:55 UTC
Buy the cheapest on eBay, $8,099.99
free shipping · New · gpuelitecom 524 · 100.0%
Price alert
One email per move, item plus shipping. Clear the price to hear about any drop. Manage alerts.
Price history
Tracking started today. The chart appears once there is a second day of data.
Where to buy
| Vendor | Price | Condition | Observed | |
|---|---|---|---|---|
| eBay cheapest active listing |
$8,099.99 free shipping |
new and used | 2026-10-18 02:55 UTC | Go to eBay |
On eBay now
3 listings passed our filters at the last check. Cheapest delivered first. Seller reputation is feedback count and positive rate.
-
$8,099.99free shipping -
$8,380free shipping -
$8,399free shipping
What it runs
For each model, grouped by vendor with the newest first: the largest quant that fits with 8K of context and the estimated speed for one user, at 8K and at 32K of context. On a card, a model too big for its memory can still run with part of the weights in system RAM; we assume 64 GB, and the speed shown accounts for it. The model you picked on the ladder is highlighted. Estimates from the spec sheet, not measurements: see how we estimate.
| Model | Quant | Needs at 8K | Fit | Max context | Decode at 8K | At 32K | Prefill tok/s |
|---|---|---|---|---|---|---|---|
| DeepSeek | |||||||
| DeepSeek V4.1 Flash 552B-A16B Sep 2026 | IQ2_M | 189 GB | fits | 128K+ | ~23 | ~23 | ~230 |
| DeepSeek V4 Flash 284B-A13B Jul 2026 | Q4_K_M | 175 GB | fits | 128K+ | ~15 | ~14 | ~290 |
| DeepSeek V3.1 671B-A37B Aug 2025 | Q4_K_M | 411 GB | does not fit | ||||
| Gemma | |||||||
| Gemma 4 12B May 2026 | Q8_0 | 14 GB | fits | 128K+ | ~11 | ~10 | ~310 |
| Gemma 4 31B Mar 2026 | Q8_0 | 35 GB | fits | 128K+ | ~4.1 | ~3.9 | ~120 |
| Gemma 4 26B-A4B Mar 2026 | Q8_0 | 28 GB | fits | 128K+ | ~29 | ~26 | ~980 |
| Gemma 3 27B Mar 2025 | Q8_0 | 31 GB | fits | 128K+ | ~4.6 | ~4.3 | ~140 |
| GLM | |||||||
| GLM-5.3 753B-A40B Aug 2026 | Q4_K_M | 462 GB | does not fit | ||||
| GLM-5.3 Flash 320B-A18B Aug 2026 | Q3_K_M | 159 GB | fits | 128K+ | ~14 | ~13 | ~210 |
| GLM-4.6 355B-A32B Sep 2025 | Q3_K_M | 179 GB | fits | 42K | ~7.1 | ~4.8 | ~120 |
| GLM-4.5 Air 106B-A12B Jul 2025 | Q8_0 | 115 GB | fits | 128K+ | ~9.3 | ~7.0 | ~310 |
| gpt-oss | |||||||
| gpt-oss-120b Aug 2025 | Q8_0 | 126 GB | fits | 128K+ | ~21 | ~16 | ~730 |
| gpt-oss-20b Aug 2025 | Q8_0 | 24 GB | fits | 128K+ | ~29 | ~23 | ~1,000 |
| Llama | |||||||
| Llama 4 Scout 109B-A17B Apr 2025 | Q8_0 | 119 GB | fits | 128K+ | ~6.8 | ~5.5 | ~220 |
| Llama 3.3 70B Dec 2024 | Q8_0 | 79 GB | fits | 128K+ | ~1.8 | ~1.6 | ~53 |
| Llama 3.1 8B Jul 2024 | Q8_0 | 11 GB | fits | 128K+ | ~14 | ~11 | ~460 |
| MiMo | |||||||
| MiMo V2.6 Pro 1.02T-A42B Sep 2026 | Q4_K_M | 624 GB | does not fit | ||||
| MiMo V2.6 Flash 309B-A15B Sep 2026 | Q4_K_M | 190 GB | fits | 70K | ~14 | ~13 | ~250 |
| Mistral | |||||||
| Mistral Medium 3.5 128B Mar 2026 | Q8_0 | 140 GB | fits | 128K+ | ~1.0 | ~0.9 | ~29 |
| Mistral Small 4 119B-A6.5B Mar 2026 | Q8_0 | 132 GB | fits | 106K | ~11 | ~5.1 | ~570 |
| Mistral Small 3.2 24B Jun 2025 | Q8_0 | 28 GB | fits | 128K+ | ~5.1 | ~4.4 | ~150 |
| Nemotron | |||||||
| Nemotron 3.5 Lightning 30B-A3B Aug 2026 | Q8_0 | 35 GB | fits | 128K+ | ~32 | ~31 | ~1,100 |
| Qwen | |||||||
| Qwen3.8 Flash-Next 125B-A6B Aug 2026 | Q8_0 | 134 GB | fits | 128K+ | ~19 | ~18 | ~620 |
| Qwen3.8 27B Aug 2026 (your pick) | Q8_0 | 31 GB | fits | 128K+ | ~4.7 | ~4.4 | ~140 |
| Qwen3.6 27B Apr 2026 | Q8_0 | 31 GB | fits | 128K+ | ~4.7 | ~4.4 | ~140 |
| Qwen3.6 35B-A3B Apr 2026 | Q8_0 | 39 GB | fits | 128K+ | ~36 | ~32 | ~1,200 |
| Qwen3.5 9B Feb 2026 | Q8_0 | 11 GB | fits | 128K+ | ~14 | ~13 | ~410 |
| Qwen3 Coder 480B-A35B Jul 2025 | IQ2_M | 167 GB | fits | 104K | ~9.4 | ~6.6 | ~110 |
| Qwen3 235B-A22B Jul 2025 | Q5_K_M | 170 GB | fits | 119K | ~7.7 | ~6.1 | ~170 |
| Qwen3 32B Apr 2025 | Q8_0 | 38 GB | fits | 128K+ | ~3.7 | ~3.1 | ~110 |
Specifications
- Memory
- 192 GB LPDDR5x-8533 unified
- Memory bandwidth
- 273 GB/s
- Bus width
- 256-bit
- Shader cores
- 2,560
- FP32
- 14.8 TFLOPS
- FP16 (dense)
- 30 TFLOPS
- Power
- 140 W
- Form factor
- Mini PC or small desktop (Framework, Minisforum, AMD Ryzen AI Halo)
- CPU
- AMD Ryzen AI Max+ PRO 495, 16 cores, 40-CU Radeon GPU
- Architecture
- Gorgon Halo
- Launched
- 2026
- MSRP
- $7,299
Source: manufacturer specifications.