Nvidia Blackwell GPU
Shares rank 25 of 36 graphics cards by recommenders
More people criticize it than recommend it. People recommend Blackwell GPUs for running local AI models quickly, with high prices and very high power draw as the main drawbacks.
7 people recommended Nvidia Blackwell GPU in Hacker News comments, 2 had mixed feelings and 9 criticized it, November 2025 to October 2026.
Summary written by Claude from the comments below.
18 people gave an opinion
- recommend it
- 7
- mixed feelings
- 2
- warn against or regret it
- 9
The record
Recommended in 7 of 12 months
In 7 different threads.
- November 2025: recommended
- December 2025: not recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: not recommended
- September 2026: recommended
- October 2026: not recommended
Why it doesn’t carry the badge
- 7 people recommend itNeeds 10 or more, not met.
- 0.7 recommenders for every criticNeeds 2 to 1 or better, not met.
- Recommended in 7 of 12 monthsNeeds 4 or more, met.
- Recommended in 7 threadsNeeds 5 or more, met.
- The biggest thread supplies 14% of its recommendersNeeds half or less, met.
The same rules apply to every product, and the badge can’t be bought. Mixed feelings count toward neither side. How the badge works
What people say about it
Reasons grouped from a sample of 14 of the 18 opinions about it, picked in proportion to how many people recommend it, have mixed feelings and criticize it. Each mark is one comment in the sample. Open a reason to read what they said.
What people warn about
Each mark is one comment
High prices and rising costs
“whose Blackwell gen prices are now up 25% from launch, with more increases to come”
Comment from 21 May 2026 in “Ask HN: Is the next big thing locally running coding agents?” “Single 96GB Blackwell" is still $15K+ worth of hardware.”
Comment from 5 Feb 2026 in “Claude Code: connect to a local model when your quota runs out” Very high power draw and electricity cost
“they haven't faced any actual competitive pressure against things like power draws in the multi-kilowatt range for the Blackwells”
Comment from 15 Jan 2026 in “Furiosa: 3.5x efficiency over H100s” “using online SOTA models would had caught it earlier without usd12k and monthly electricity being spent”
Comment from 18 Jun 2026 in “Local Qwen isn't a worse Opus, it's a different tool” Small uplift over previous generation
“Blackwell was already kind of a dud on performance uplift from Ada beyond the new LLM features”
Comment from 12 Sep 2026 in “Nvidia is the central bank of AI” “Blackwell is the last generation Nvidia didn't optimize specifically for inference.”
Comment from 4 Jul 2026 in “GLM5.2 on AMD MI355X at 2626 tok/s/node at over 2x lower cost than Blackwell”
Why people recommend it
Each mark is one comment
Runs local AI models well
“If you spend twice what I did and go Nvidia you should have nearly no issues running any models.”
Comment from 12 Jan 2026 in “Anthropic: Developing a Claude Code competitor using Claude Code is banned” “for inference the difference widens a bunch makes a lot more sense to go with Blackwell”
Comment from 11 Sep 2026 in “Thelio Mira AI Linux Workstation: 192 GB GPU Memory” “it is pretty cool to get beaten in a programming contest by my own video card”
Comment from 5 Feb 2026 in “Claude Code: connect to a local model when your quota runs out” - Comment from 6 Jun 2026 in “Nvidia is proposing a beast of a CPU system for Windows PCs”
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 | Mixed | Criticize | Approval | Months | Had for | Badge |
|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 3090 | 88 | 26 | 18 | 83% | 12 | 4 yrs11 people | Approved |
| NVIDIA GeForce RTX 5090 | 73 | 26 | 34 | 68% | 12 | — | Approved |
| NVIDIA RTX PRO 6000 | 58 | 15 | 19 | 75% | 12 | — | Approved |
| AMD Radeon GPU | 43 | 4 | 29 | 60% | 11 | — | — |
| Nvidia Blackwell GPUThis page | 7 | 2 | 9 | 44% | 7 | — | — |
Also compared with Nvidia GPUs, AMD Radeon AI Pro R9700, Intel Arc Pro B70 and 3dfx Voodoo.
What people said
14 of the 18 opinions about it, picked in proportion to the three counts, newest first. Each links to the original comment.
“Blackwell was already kind of a dud on performance uplift from Ada beyond the new LLM features”
“any model over about 128GB of weights isn't going to fit on the GPUs”
“for inference the difference widens a bunch makes a lot more sense to go with Blackwell”
“Blackwell is the last generation Nvidia didn't optimize specifically for inference.”
“using online SOTA models would had caught it earlier without usd12k and monthly electricity being spent”
“spit tokens 10-20x the speed”
“having a blackwell graphics card I can evalute with linux and windows, both for ML training, inference, and gaming, is a huge network effect”
“whose Blackwell gen prices are now up 25% from launch, with more increases to come”
“it is pretty cool to get beaten in a programming contest by my own video card”
“Single 96GB Blackwell" is still $15K+ worth of hardware.”
“they haven't faced any actual competitive pressure against things like power draws in the multi-kilowatt range for the Blackwells”
“If you spend twice what I did and go Nvidia you should have nearly no issues running any models.”
“Nvidia may have the superior architecture at the single-chip level”
“I'm going to be a lot wealthier than the Nvidia bears.”
Counts come from 1,791,911 Hacker News comments labeled by Claude, November 2025 to October 2026, with data through October 31, 2026. Counts updated 2 Nov 2026. How we count. Report a mistake.