NVIDIA RTX A6000
Shares rank 29 of 42 graphics cards by recommenders
Fast for running local AI models, and people report two-card builds working well. Common caveats: it is loud, draws about 500W, and its memory is limited.
Fast for running local AI models, and people report two-card builds working well. Common caveats: it is loud, draws about 500W, and its memory is limited.
8 people gave an opinion
- recommend it
- 8
- criticize it
- 0
- approval share
- 100%
Recommended in 4 of 13 months
- November 2025: recommended
- December 2025: not recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: not recommended
- June 2026: not recommended
- July 2026: not recommended
- August 2026: not recommended
- September 2026: recommended
- October 2026: not recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 8 people recommend itNeeds 10 or more, not met.
- Nobody criticizes itNeeds 2 to 1 or better, met.
- Recommended in 4 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Fast for local models
“Wow that thing was fast but noisy and also burns 500W”
Comment from 15 Jan 2026 in “Ask HN: How are you doing RAG locally?” “Two A6000 is fast but quite limited in memory.”
Comment from 16 Feb 2026 in “Qwen3.5: Towards Native Multimodal Agents” Works in two-GPU builds
“We ended up picking a ~$12,000 computer with 2 A6000s.”
Comment from 4 Feb 2026 in “Claude Code: connect to a local model when your quota runs out” “To 0 people's surprise, the 2xA6000s did work.”
Comment from 5 Feb 2026 in “Claude Code: connect to a local model when your quota runs out” “I proposed A6000. That ended up working.”
Comment from 2 Sep 2026 in “My local model setup on an M4 Pro Mac Mini” Good value versus newer cards
“Maybe two/four of the older A100/A6000 could do the trick too.”
Comment from 10 Nov 2025 in “Asus Ascent GX10” “they aren't that bad compared to a single RTX 6000 Pro”
Comment from 4 Feb 2026 in “FlashAttention-T: Towards Tensorized Attention” “For the valuable kick start usecase it pays off.”
Comment from 23 Jan 2026 in “What Will You Do When AI Will Run Out of Money and Disappear?”
What people warn about
Each mark is one comment
Loud and power hungry
“Wow that thing was fast but noisy and also burns 500W”
Comment from 15 Jan 2026 in “Ask HN: How are you doing RAG locally?” Limited memory for large models
“Two A6000 is fast but quite limited in memory.”
Comment from 16 Feb 2026 in “Qwen3.5: Towards Native Multimodal Agents”
Other graphics cards people recommend
The four most recommended other graphics cards, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 3090 | 116 | 21 | 85% | 12 | 5 yrs14 owners | Approved |
| NVIDIA GeForce RTX 5090 | 102 | 37 | 73% | 12 | 1 yr3 owners | Approved |
| NVIDIA RTX PRO 6000 | 77 | 26 | 75% | 12 | — | Approved |
| Nvidia GPUs | 67 | 67 | 50% | 11 | 6 yrs3 owners | — |
| NVIDIA RTX A6000This page | 8 | 0 | 100% | 4 | — | — |
What people said
All 9 comments that mention it, newest first. Each links to the original comment.
“I proposed A6000. That ended up working.”
“Two A6000 is fast but quite limited in memory.”
“To 0 people's surprise, the 2xA6000s did work.”
“We ended up picking a ~$12,000 computer with 2 A6000s.”
“they aren't that bad compared to a single RTX 6000 Pro”
“For the valuable kick start usecase it pays off.”
“Wow that thing was fast but noisy and also burns 500W”
“also runs the 120b just fine (quantized)”
“Maybe two/four of the older A100/A6000 could do the trick too.”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.