Kingy AI · Local AI hardware verdicts · Standard v0.1
Can You Run It at Home?
Kingy’s verdict for each model on six common machine types, from a 16GB laptop to a 512GB Mac Studio. Every verdict names the exact version and says how Kingy knows: measured, reported by a named third party, or estimated.
Estimated verdicts are calculated, not run. Kingy works them out from published file sizes and memory bandwidth, and labels them “(estimated)”. A measured or reported result always wins over an estimate.
24 verdicts for 4 models: 1 Kingy measured · 10 Reported · 10 Estimated · 3 No evidence yet. Data updated Oct 18, 2026.
Reference machine: RTX 4090 24GB with 64GB DDR5 system RAM · 22.5 GB VRAM (24 minus 1.5 for driver and display) usable · 1,008 GB/s GPU · ~80 GB/s system RAM. Best verdict first.
Qwen3.8-27B
Text model · Qwen (Alibaba)
Can you run Qwen3.8-27B at home? Yes. PC with a 24GB GPU · Q4_K_M · llama.cpp b10453, commit 3cb7ffb · 49.1 tok/s · Kingy measured Aug 17, 2026
Yes
Kingy measured 49.1 tok/s in a synthetic decode test (llama-bench) on an RTX 4090 desktop. 8K, 16K, 32K and 64K contexts all ran fully on the GPU; 64K peaked at 21.25 GB, under 22.5 GB. Longer contexts were not tested.
Can you run MiniMax H3 at home? Barely. PC with a 24GB GPU · FP8 scaled, pruned (ComfyUI build) · ComfyUI 0.30.0 · 5.5 min per 5 s clip at 1344x768 (124 frames), cold start · Reported Aug 7, 2026
Barely
Barely. A 5 s 1344x768 clip took about 330 s cold (300 s warm) with the turbo LoRA at 12 steps, but the reported VRAM peak (15.7 GB) was below the 21 GB transformer file, so part of it ran from system RAM. Reported, not Kingy-tested.
Can you run Qwen3.8-Flash-Next at home? Barely (estimated). PC with a 24GB GPU · GGUF Q2_0 (GSQ-RCO) · llama.cpp mainline with qwen4exp · 22.2 tok/s at 8K context · Estimated Oct 18, 2026
Barely (estimated)
Barely (estimated). Only a ~2.4-bit community GGUF fits 22.5 GB VRAM plus 64 GB RAM, with about 70% of the weights in system RAM. Worst-case estimate: 22 tok/s at 8K. The official FP8 files need about 205 GB.
Can you run DeepSeek V4.1 Flash 1017 at home? No (estimated). PC with a 24GB GPU · VQ GGUF (8-weight vector quant) · Estimated Oct 18, 2026
No (estimated)
No (estimated). The smallest published 1017 build, a 113.6GB VQ base, is more than the RTX 4090 reference PC's 22.5GB VRAM plus 64GB RAM (86.5GB), before the Engram tables. Third-party SSD-streaming runs use 128GB RAM, so they are not this machine.
Every verdict follows Kingy’s “Can you run it at home?” standard v0.1. A verdict rates one model on one machine type, names the exact version, and always says how Kingy knows.
Text models
Verdict
What it takes
Yes
Loads entirely in GPU or unified memory, generates at least 20 tokens/s, works at 8K context or more, and starts answering an 8K prompt within 30 seconds. All four must hold.
Yes, but slow
Loads and runs, but generates 5–20 tokens/s, or takes more than 30 seconds to start answering an 8K prompt.
Barely
Runs, but generates under 5 tokens/s, or needs more than 25% of the weights in system RAM (weights read from SSD count too), or fits only at 3-bit or lower, or only with less than 4K context.
No
Will not load on that machine at any published version.
Speed is generation speed for one user (batch 1) at the stated context.
Image and video models
Kingy times one standard output: a 5-second clip at the model’s native resolution, or one 1024×1024 image.
Verdict
One 5-second clip
One image
Yes
5 min or less
30 s or less
Yes, but slow
5–30 min
30 s to 3 min
Barely
Over 30 min, only at reduced resolution or length, or with heavy offload (ComfyUI partial offload counts)
Over 3 min, or with offload
No
Won’t load
Won’t load
Media speed depends too much on compute and pipeline settings to estimate, so media verdicts are always measured or reported. Without either, they read “Not tested yet”.
How Kingy knows
Kingy measured: Kingy ran it. Each result has a Local Lab receipt with the exact file and hash, runtime build, command, context, speed and date. Failed attempts are kept too.
Reported: a named third party ran it. Kingy links the report and records their machine, file, runtime and speed. “Runs fine” without numbers does not count.
Estimated: calculated, not run, from the published file size and the formula below. An estimate always says “(estimated)” and never shows as a bare “Yes”.
No evidence yet: no measurement, no reliable report and, for text models, no published weight size to estimate from. The verdict is then “Not tested yet”. Kingy never invents a verdict to avoid saying so.
It fits if the memory needed is no more than the machine type’s usable memory.
Speed ≈ 0.5 × memory bandwidth ÷ bytes read per token, where bytes per token = active parameters × bits per weight ÷ 8. Mixture-of-experts models use active, not total, parameters. Real runtimes usually reach 60–80% of bandwidth, so 0.5 errs on the slow side.
On GPU PCs, if more than 25% of the weights must sit in system RAM, the estimate is Barely at best.
An estimated “Yes” needs at least 40 tokens/s, twice the threshold, and 20% spare memory. Otherwise the best an estimate can say is “Yes, but slow”. An estimated “No” means even the smallest published version is bigger than the machine’s usable memory plus, on GPU PCs, the system RAM. When a measurement and an estimate disagree, the measurement wins.
Machine types
Verdicts are per machine type, not per exact computer. Each type has a reference machine, and estimates use its numbers. Newer chips are often faster, so estimates err on the slow side. GB means 10⁹ bytes.
Machine type
Reference machine
Memory a model can use
Memory bandwidth
16GB laptop
MacBook Air M4, 16GB
10.5 GB (macOS lets the GPU use about 2/3 of RAM)
120 GB/s
PC with a 24GB GPU
RTX 4090 24GB with 64GB DDR5 system RAM
22.5 GB VRAM (24 minus 1.5 for driver and display)
Every verdict names the file tested or estimated, its format, the runtime, and how it compares with the official release. Local does not automatically mean compressed, and online does not automatically mean full quality.
Measured and reported verdicts are marked “Recheck due” after 90 days, estimates after 30 days. The old result stays visible until the recheck.
Use the data
The same records feed the verdict box in each Kingy post, the can_i_run_locally tool on the Kingy Facts MCP server, and the public API: