Apple · M1 Ultra · 128 GB · Qwen3.8 27B · Q4_K_M at 8K context: ~28 tok/s
Apple Mac Studio M1 Ultra, 128 GB
The first Mac Studio with 128 GB, from 2022: the same 800 GB/s as the M2 Ultra with a slower GPU, so decode speed is close and prefill is slower. A 70B model at Q8 fits. Usually the cheapest 128 GB machine at this bandwidth.
Lowest price right now
$4,000
eBay · free shipping · $31 per GB · observed 2026-11-14 18:37 UTC
Buy the cheapest on eBay, $4,000
free shipping · Used · dingledodies 184 · 100.0%
Price alert
One email per move, item plus shipping. Clear the price to hear about any drop. Manage alerts.
Price history
Daily low across every tracked vendor, last 4 days.
Where to buy
| Vendor | Price | Condition | Observed | |
|---|---|---|---|---|
| eBay cheapest active listing |
$4,000 free shipping |
new and used | 2026-11-14 18:37 UTC | Go to eBay |
On eBay now
9 listings passed our filters at the last check, 7 min ago. We check every 28 minutes. Cheapest delivered first. Seller reputation is feedback count and positive rate.
-
$4,000free shipping -
$4,099free shipping -
$4,099free shipping -
$4,160free shipping -
$4,399free shipping -
$4,599.99free 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 | IQ2_M | 98 GB | fits | 128K+ | ~74 | ~62 | ~150 |
| 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+ | ~37 | ~36 | ~170 |
| Gemma 4 31B Mar 2026 | Q8_0 | 35 GB | fits | 128K+ | ~14 | ~14 | ~65 |
| Gemma 4 26B-A4B Mar 2026 | Q8_0 | 28 GB | fits | 128K+ | ~82 | ~75 | ~520 |
| Gemma 3 27B Mar 2025 | Q8_0 | 31 GB | fits | 128K+ | ~16 | ~15 | ~73 |
| GLM | |||||||
| GLM-5.3 753B-A40B Aug 2026 | Q4_K_M | 462 GB | does not fit | ||||
| GLM-5.3 Flash 320B-A18B Aug 2026 | IQ2_M | 111 GB | fits | 128K+ | ~59 | ~52 | ~110 |
| GLM-4.6 355B-A32B Sep 2025 | IQ2_M | 125 GB | fits | 15K | ~31 | ~62 | |
| GLM-4.5 Air 106B-A12B Jul 2025 | Q8_0 | 115 GB | fits | 73K | ~30 | ~23 | ~170 |
| gpt-oss | |||||||
| gpt-oss-120b Aug 2025 | Q8_0 | 126 GB | fits | 35K | ~62 | ~51 | ~390 |
| gpt-oss-20b Aug 2025 | Q8_0 | 24 GB | fits | 128K+ | ~81 | ~68 | ~550 |
| Llama | |||||||
| Llama 4 Scout 109B-A17B Apr 2025 | Q8_0 | 119 GB | fits | 54K | ~23 | ~18 | ~120 |
| Llama 3.3 70B Dec 2024 | Q8_0 | 79 GB | fits | 128K+ | ~6.2 | ~5.6 | ~28 |
| Llama 3.1 8B Jul 2024 | Q8_0 | 11 GB | fits | 128K+ | ~50 | ~38 | ~250 |
| 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 | IQ2_M | 107 GB | fits | 128K+ | ~69 | ~62 | ~130 |
| Mistral | |||||||
| Mistral Medium 3.5 128B Mar 2026 | Q6_K | 109 GB | fits | 58K | ~4.4 | ~4.1 | ~16 |
| Mistral Small 4 119B-A6.5B Mar 2026 | Q6_K | 104 GB | fits | 47K | ~41 | ~18 | ~310 |
| Mistral Small 3.2 24B Jun 2025 | Q8_0 | 28 GB | fits | 128K+ | ~18 | ~16 | ~83 |
| Nemotron | |||||||
| Nemotron 3.5 Lightning 30B-A3B Aug 2026 | Q8_0 | 35 GB | fits | 128K+ | ~88 | ~86 | ~570 |
| Qwen | |||||||
| Qwen3.8 Flash-Next 125B-A6B Aug 2026 | Q6_K | 104 GB | fits | 128K+ | ~71 | ~65 | ~330 |
| Qwen3.8 27B Aug 2026 (your pick) | Q8_0 | 31 GB | fits | 128K+ | ~16 | ~16 | ~74 |
| Qwen3.6 27B Apr 2026 | Q8_0 | 31 GB | fits | 128K+ | ~16 | ~16 | ~74 |
| Qwen3.6 35B-A3B Apr 2026 | Q8_0 | 39 GB | fits | 128K+ | ~95 | ~87 | ~660 |
| Qwen3.5 9B Feb 2026 | Q8_0 | 11 GB | fits | 128K+ | ~49 | ~45 | ~220 |
| Qwen3 Coder 480B-A35B Jul 2025 | Q4_K_M | 296 GB | does not fit | ||||
| Qwen3 235B-A22B Jul 2025 | Q3_K_M | 118 GB | fits | 57K | ~34 | ~26 | ~90 |
| Qwen3 32B Apr 2025 | Q8_0 | 38 GB | fits | 128K+ | ~13 | ~11 | ~61 |
Specifications
- Memory
- 128 GB LPDDR5 unified
- Memory bandwidth
- 800 GB/s
- Bus width
- 1024-bit
- FP32
- 15.9 TFLOPS
- FP16 (dense)
- 16 TFLOPS
- Form factor
- Desktop box, 197 x 197 x 95 mm
- CPU
- 20-core CPU, 48-core GPU (64-core optional), 32-core Neural Engine
- Architecture
- M1 Ultra
- Launched
- 2022
- MSRP
- $4,799
Source: manufacturer specifications.