Hardware → models
RTX 6000 Ada 48 GB
Start with gpt-oss-20b at MXFP4 profile for general chat. For a coding-focused option, try Qwen3-Coder 30B-A3B Instruct.
Models to start with
Planning estimates48 GB dedicated VRAM · 4,000 tokens · 32 GB system RAM · full GPU residency
| Model / starting file | Why consider it | GPU memory estimate | Memory left¹ |
|---|---|---|---|
| gpt-oss-20bMXFP4 profile | Reasoning and tool-oriented tasksNeeds a runtime that supports MXFP4 and the Harmony conversation format. | 14.8 GB | 31.8 GB |
| Qwen3-Coder 30B-A3B InstructQ4_K_M | Coding mixture-of-experts modelFull weights must fit despite 3.3B active parameters. Use a compatible runtime. | 20.8 GB | 25.8 GB |
| Qwen3 32BQ4_K_M | Larger general-purpose modelMore memory than 14B. Keep thinking mode fixed when comparing answers. | 21.6 GB | 25.0 GB |
| Llama 3.2 3B InstructQ4_K_M | Small text assistantA smaller weight footprint. Check the exact Llama checkpoint and licence. | 3.7 GB | 42.9 GB |
From the published model guides here; the Finder covers the full catalogue. ¹ Memory left after the Finder’s reserve. Other applications can use this space. Larger models and higher precision do not guarantee better answers.
This is the 48 GB Ada Generation card. The larger memory budget does not establish speed or better output quality. CPU offloading changes the memory allocation and may slow generation.
Setup and common questions
Is this the same as an RTX A6000?
No. This page covers RTX 6000 Ada Generation, with 48 GB GDDR6 ECC and Ada architecture. RTX A6000 and RTX PRO 6000 are separate products. Select the exact card in the Finder rather than matching only the name.
Should I immediately choose the largest model?
More VRAM expands the available memory budget; it does not establish better answers or quicker inference. Compare a larger model with a smaller one on the same task. The shortlist is limited to the reviewed page subjects; the Finder covers more models.
Is 32 GB system RAM enough?
That is the assumption for the profiles shown, not a universal workstation recommendation. Large files, other applications, conversion and CPU offloading may need more. Check the exact runtime’s peak host memory before planning a larger setup.
How are these recommendations chosen?
We use the main Finder’s balanced task selection and memory calculation, limited to models with published guides here. Task scores and speed proxies are planning inputs, not measured quality rankings.
GPU memory includes weights, context cache and a 0.9 GB runtime allowance, plus a 3% reserve (0.15–1.5 GB). Cache format and runtime buffers can change the fit. See the calculation assumptions.
Sources and related guides
Specification sources checked 3 October 2026.
These comparisons are planning estimates. No reviewed or independently reproduced measurements are available yet. About test evidence.