Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Geeky Gadgets reports that a custom PC with an RTX 3090 generated sustained output at 140 tokens per second, compared with 85 tokens per second on a 64GB Mac Studio with M5 Max. That is a notable result for the PC in the comparison it describes—not proof that every RTX 3090 will be faster than every Mac Studio. The summary does not disclose enough test details to establish a controlled, repeatable head-to-head benchmark.
What the reported benchmark measured
The September 29, 2026 Geeky Gadgets summary of a comparison by The Stack names Qwen 3.6, described in the article as a 35-billion-parameter model. It reports these results:
As an Amazon Associate I earn from qualifying purchases.
| Measure | RTX 3090 PC | 64GB M5 Max Mac Studio |
|---|---|---|
| Sustained generated output | 140 tokens per second, as reported by Geeky Gadgets in 2026 | 85 tokens per second, as reported by Geeky Gadgets in 2026 |
| Short-prompt processing | About 3,000 tokens per second, as reported by Geeky Gadgets in 2026 | About 3,000 tokens per second, as reported by Geeky Gadgets in 2026 |
| Longer-prompt processing | not stated in the Geeky Gadgets summary | About 2,000 tokens per second, as reported by Geeky Gadgets in 2026 |
| Context described as usable at full speed | About 90,000–150,000 tokens, according to the summary | 262,000 tokens, according to the summary |
Prompt processing and generated output are different stages, so their tokens-per-second figures should not be treated as one interchangeable speed score. The summary does not identify the exact test harness, quantization, runtime versions, batch size, thermal conditions, or whether both machines used identical model artifacts. The linked Stack video page does not provide accessible methodology details that resolve those questions.
Why 140 tokens per second is not a universal RTX 3090 result
Local model speed depends on more than the graphics card name: model implementation, quantization, context length, and software all matter. Since the source summary does not establish parity on these factors, its 140-versus-85 result is best read as the outcome reported for those particular setups and test conditions, not as a guaranteed ratio for another computer or workload.
#1 Best Overall
- Item Package Dimension - 15.0L x 12.25W x 4.25H inches
- Item Package Weight - 6.0 Pounds
- Item Package Quantity - 1
- Product Type - VIDEO CARD
Memory and context are a separate trade-off
The comparison describes the Mac handling a 262,000-token context window, while the RTX setup handles about 90,000 to 150,000 tokens at full speed. Those are claims about the setups in the summary, not general limits for every model or runtime.
The machines’ memory figures also are not directly interchangeable. Apple lists M5 Max Mac Studio configurations with 48GB, 64GB, or 128GB of unified memory on its technical specifications page. The comparison’s PC uses a discrete RTX 3090 and system RAM; the article describes a 24GB card and says the PC has 32GB of RAM. A GPU’s memory and a computer’s unified memory do not automatically provide equivalent capacity or performance for a given model. Geeky Gadgets says 32GB of system RAM is generally sufficient for the Qwen workload it describes and that 128GB adds little in that setup; that is not a broad recommendation for other models.
What the comparison says about cost
Geeky Gadgets gives approximate prices of $2,000 for its custom RTX 3090 PC and $3,799 for the 64GB Mac Studio. These are the article’s 2026 estimates, not verified current quotes. A PC’s total depends on its parts and, in particular, the condition and price of the graphics card; the comparison’s figures should not be used as a current shopping budget without checking listings.
Which setup fits your local AI priorities?
Choose the RTX 3090 PC if sustained generation speed is the priority
In the reported comparison, the PC has the higher sustained output rate. A configurable desktop may also suit someone prepared to select parts and manage a custom setup. The disclosed results do not establish how another RTX 3090 build will perform, so compare tests using your model, quantization, runtime, and typical context length before committing.
Rank #3
- Digital Maximum Resolution - 7680 X 4320
- Output- Displayport X 3 (V1.4A) / Hdmi 2.1 X 1
- Memory Interface- 384-Bit
- Package Quantity-1
Choose the Mac Studio if larger context and an integrated system matter more
The comparison reports a larger usable context for the 64GB Mac Studio, alongside lower sustained generation speed. Apple’s M5 Max family includes several unified-memory configurations, so confirm the specific configuration rather than assuming all Mac Studios have 64GB.
For a fair purchase decision, match the workload first: model and quantization, prompt length, desired context, and whether you care more about prompt processing or generated output. Then seek results for those same conditions; the published comparison does not disclose enough to make its figures a universal GPU-versus-Apple-Silicon ranking.
Quick Recap
Best Value
- Memory Speed:19.5 Gbps.Digital Max Resolution:7680 x 4320
- NVIDIA Ampere Streaming Multiprocessors: The building blocks for the world’s fastest, most efficient GPU, the all-new Ampere SM brings 2X the FP32 throughput and improved power efficiency.
- 2nd Generation RT Cores: Experience 2X the throughput of 1st gen RT Cores, plus concurrent RT and shading for a whole new level of ray tracing performance.
- 3rd Generation Tensor Cores: Get up to 2X the throughput with structural sparsity and advanced AI algorithms such as DLSS. Now with support for up to 8K resolution, these cores deliver a massive boost in game performance and all-new AI capabilitiesAvoid using unofficial software
- Axial-Tech Fan Design has been newly tuned with a reversed central fan direction for less turbulence.
Rank #4
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Recommended Free Tools
