Apple · M3 Ultra · 256 GB · Qwen3.8 27B · Q4_K_M: ~30 tok/s
Apple Mac Studio M3 Ultra, 256 GB
256 GB at 819 GB/s: a 235B mixture-of-experts model at Q4, or a 70B at FP16, in a silent box. Half the price of the 512 GB configuration.
Lowest price right now
$8,499
eBay · free shipping · $33 per GB · observed 2026-10-14 22:19 UTC
Where to buy
| Vendor | Price | Condition | Observed | |
|---|---|---|---|---|
| eBay cheapest active listing |
$8,499 free shipping |
new and used | 2026-10-14 22:19 UTC | Go to eBay |
| Amazon price tracking coming |
price on Amazon | new | Go to Amazon |
On eBay now
25 listings passed our filters at the last check. Cheapest delivered first. Seller reputation is feedback count and positive rate.
-
$8,499free shipping -
$8,499free shipping -
$8,650plus shipping -
$8,816.98$8,750 + $66.98 shipping -
$8,999free shipping -
$8,999free 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. 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 tok/s | Prefill tok/s |
|---|---|---|---|---|---|---|
| DeepSeek | ||||||
| DeepSeek V4.1 Flash 552B-A16B Sep 2026 | IQ2_M | 189 GB | fits | 128K+ | ~90 | ~170 |
| DeepSeek V4 Flash 284B-A13B Jul 2026 | Q6_K | 235 GB | fits | 128K+ | ~46 | ~200 |
| DeepSeek V3.1 671B-A37B Aug 2025 | IQ2_M | 230 GB | fits | 128K+ | ~39 | ~72 |
| Gemma | ||||||
| Gemma 4 12B May 2026 | Q8_0 | 14 GB | fits | 128K+ | ~39 | ~220 |
| Gemma 4 31B Mar 2026 | Q8_0 | 35 GB | fits | 128K+ | ~15 | ~87 |
| Gemma 4 26B-A4B Mar 2026 | Q8_0 | 28 GB | fits | 128K+ | ~120 | ~700 |
| Gemma 3 27B Mar 2025 | Q8_0 | 31 GB | fits | 128K+ | ~17 | ~97 |
| GLM | ||||||
| GLM-5.3 753B-A40B Aug 2026 | Q4_K_M | 462 GB | does not fit | |||
| GLM-5.3 Flash 320B-A18B Aug 2026 | Q5_K_M | 229 GB | fits | 128K+ | ~38 | ~150 |
| GLM-4.6 355B-A32B Sep 2025 | Q4_K_M | 221 GB | fits | 98K | ~25 | ~83 |
| GLM-4.5 Air 106B-A12B Jul 2025 | Q8_0 | 115 GB | fits | 128K+ | ~39 | ~220 |
| gpt-oss | ||||||
| gpt-oss-120b Aug 2025 | Q8_0 | 126 GB | fits | 128K+ | ~91 | ~520 |
| gpt-oss-20b Aug 2025 | Q8_0 | 24 GB | fits | 128K+ | ~130 | ~740 |
| Llama | ||||||
| Llama 4 Scout 109B-A17B Apr 2025 | Q8_0 | 119 GB | fits | 128K+ | ~27 | ~160 |
| Llama 3.3 70B Dec 2024 | Q8_0 | 79 GB | fits | 128K+ | ~6.6 | ~38 |
| Llama 3.1 8B Jul 2024 | Q8_0 | 11 GB | fits | 128K+ | ~58 | ~330 |
| 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 | Q6_K | 255 GB | fits | 39K | ~40 | ~180 |
| Mistral | ||||||
| Mistral Medium 3.5 128B Mar 2026 | Q8_0 | 140 GB | fits | 128K+ | ~3.6 | ~21 |
| Mistral Small 4 119B-A6.5B Mar 2026 | Q8_0 | 132 GB | fits | 128K+ | ~71 | ~410 |
| Mistral Small 3.2 24B Jun 2025 | Q8_0 | 28 GB | fits | 128K+ | ~19 | ~110 |
| Qwen | ||||||
| Qwen3.8 Flash-Next 125B-A6B Aug 2026 | Q8_0 | 134 GB | fits | 128K+ | ~77 | ~440 |
| Qwen3.8 27B Aug 2026 (your pick) | Q8_0 | 31 GB | fits | 128K+ | ~17 | ~99 |
| Qwen3.6 27B Apr 2026 | Q8_0 | 31 GB | fits | 128K+ | ~17 | ~99 |
| Qwen3.6 35B-A3B Apr 2026 | Q8_0 | 39 GB | fits | 128K+ | ~150 | ~890 |
| Qwen3 Coder 480B-A35B Jul 2025 | Q3_K_M | 239 GB | fits | 74K | ~29 | ~76 |
| Qwen3 235B-A22B Jul 2025 | Q8_0 | 252 GB | fits | 27K | ~21 | ~120 |
| Qwen3 32B Apr 2025 | Q8_0 | 38 GB | fits | 128K+ | ~14 | ~81 |
Specifications
- Memory
- 256 GB LPDDR5 unified
- Memory bandwidth
- 819 GB/s
- Bus width
- 1024-bit
- FP32
- 21.3 TFLOPS
- FP16 (dense)
- 21 TFLOPS
- Form factor
- Desktop box, 197 x 197 x 95 mm
- CPU
- 28-core CPU, 60-core GPU (80-core optional), 32-core Neural Engine
- Architecture
- M3 Ultra
- Launched
- 2025
- MSRP
- $5,499
Source: manufacturer specifications.