AMD · RDNA 4 · 16 GB · Qwen3.8 27B · Q4_K_M: ~9.3 tok/s with RAM offload
AMD Radeon RX 9070 XT 16GB
The same RDNA 4 chip as the Radeon AI PRO R9700 with half the memory, at half the price or less. Good ROCm and Vulkan support now, but 16 GB limits it to 14B models, or mixture-of-experts models with part of the weights in system RAM.
Lowest price right now
$799.97
eBay · free shipping · $49 per GB · observed 2026-10-18 02:55 UTC
Buy the cheapest on eBay, $799.97
free shipping · Open box · uventure 8,860 · 99.7%
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 |
$799.97 free shipping |
new and used | 2026-10-18 02:55 UTC | Go to eBay |
On eBay now
98 listings passed our filters at the last check. Cheapest delivered first. Seller reputation is feedback count and positive rate.
-
$799.97free shipping -
$799.99free shipping -
$808.45$799 + $9.45 shipping -
$813.99free shipping -
$814.27$800 + $14.27 shipping -
$815.57$799 + $16.57 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 | Q4_K_M | 338 GB | does not fit | ||||
| DeepSeek V4 Flash 284B-A13B Jul 2026 | Q4_K_M | 175 GB | does not fit | ||||
| 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 | 109K | ~25 | ~24 | ~1,000 |
| Gemma 4 31B Mar 2026 | IQ2_M | 13 GB | fits | 48K | ~29 | ~24 | ~390 |
| Gemma 4 26B-A4B Mar 2026 | Q3_K_M | 14 GB | fits | 102K | ~100 | ~88 | ~3,200 |
| Gemma 3 27B Mar 2025 | Q3_K_M | 16 GB | fits | 12K | ~23 | ~440 | |
| GLM | |||||||
| GLM-5.3 753B-A40B Aug 2026 | Q4_K_M | 462 GB | does not fit | ||||
| GLM-5.3 Flash 320B-A18B Aug 2026 | Q4_K_M | 197 GB | does not fit | ||||
| GLM-4.6 355B-A32B Sep 2025 | Q4_K_M | 221 GB | does not fit | ||||
| GLM-4.5 Air 106B-A12B Jul 2025 | Q4_K_M | 68 GB | runs with 52 GB in system RAM | ~6.4 | ~510 | ||
| gpt-oss | |||||||
| gpt-oss-120b Aug 2025 | Q3_K_M | 59 GB | runs with 44 GB in system RAM | ~19 | ~16 | ~1,200 | |
| gpt-oss-20b Aug 2025 | Q4_K_M | 15 GB | fits | 33K | ~86 | ~65 | ~3,400 |
| Llama | |||||||
| Llama 4 Scout 109B-A17B Apr 2025 | Q4_K_M | 70 GB | runs with 54 GB in system RAM | ~4.5 | ~360 | ||
| Llama 3.3 70B Dec 2024 | Q4_K_M | 47 GB | runs with 32 GB in system RAM | ~1.2 | ~1.0 | ~86 | |
| Llama 3.1 8B Jul 2024 | Q8_0 | 11 GB | fits | 44K | ~33 | ~25 | ~1,500 |
| 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 | does not fit | ||||
| Mistral | |||||||
| Mistral Medium 3.5 128B Mar 2026 | Q3_K_M | 67 GB | runs with 52 GB in system RAM | ~0.8 | ~47 | ||
| Mistral Small 4 119B-A6.5B Mar 2026 | Q3_K_M | 65 GB | runs with 49 GB in system RAM | ~12 | ~930 | ||
| Mistral Small 3.2 24B Jun 2025 | Q3_K_M | 15 GB | fits | 16K | ~24 | ~510 | |
| Nemotron | |||||||
| Nemotron 3.5 Lightning 30B-A3B Aug 2026 | IQ2_M | 12 GB | fits | 128K+ | ~140 | ~130 | ~3,500 |
| Qwen | |||||||
| Qwen3.8 Flash-Next 125B-A6B Aug 2026 | Q3_K_M | 63 GB | runs with 47 GB in system RAM | ~16 | ~16 | ~1,000 | |
| Qwen3.8 27B Aug 2026 (your pick) | Q3_K_M | 15 GB | fits | 18K | ~23 | ~450 | |
| Qwen3.6 27B Apr 2026 | Q3_K_M | 15 GB | fits | 18K | ~23 | ~450 | |
| Qwen3.6 35B-A3B Apr 2026 | IQ2_M | 14 GB | fits | 123K | ~140 | ~110 | ~4,000 |
| Qwen3.5 9B Feb 2026 | Q8_0 | 11 GB | fits | 128K+ | ~33 | ~30 | ~1,300 |
| Qwen3 Coder 480B-A35B Jul 2025 | Q4_K_M | 296 GB | does not fit | ||||
| Qwen3 235B-A22B Jul 2025 | Q4_K_M | 146 GB | does not fit | ||||
| Qwen3 32B Apr 2025 | IQ2_M | 15 GB | fits | 12K | ~24 | ~370 | |
Specifications
- Memory
- 16 GB GDDR6
- Memory bandwidth
- 640 GB/s
- Bus width
- 256-bit
- Shader cores
- 4,096
- FP32
- 48.7 TFLOPS
- FP16 (dense)
- 97 TFLOPS
- Power
- 304 W
- Form factor
- Dual or triple slot, full height
- Interface
- PCIe 5.0 x16
- Architecture
- RDNA 4
- Launched
- 2025
- MSRP
- $599
Source: manufacturer specifications.