Can I run Nanbeige4.3-33B on an RTX 5090?
Key facts
Yes. Nanbeige4.3-33B runs on an RTX 5090, about 61 tokens per second (tok/s).
It needs about 23.3 GB at standard quality (estimate). Checked . bartowski Q4_K_M GGUF.
Updated · Nanbeige4.3-33B on every card
How much memory it needs
It uses about 74% of the memory that is free on this card.
Get a 1-line alert when a faster quant for Nanbeige4.3-33B drops — or when GPUs that run it drop >10%
Get a 1-line alert when a new open-weight model fits 24 GB VRAM (23.3 GB at Q4 for Nanbeige4.3-33B) or when used RTX 3090 / 4090 & Mac Studio prices drop >10% — weekly, zero spam.
What an RTX 5090 costs
GeForce RTX 5090 32GB$6,816€7,367new (lowest in-stock EU)
EU shops · Shown to visitors in Europe.US shops · Shown to visitors outside Europe.
We never pass off an estimate as a benchmark. Commissions never change which hardware we recommend. How we calculate
Price details
$6,816€7,367
How we priced this
New price, the lowest in stock in the EU.
Source: PriceSquirrel, checked 21 Oct 2026.
Cheapest in-stock listing on 21 Oct 2026: €7,367 for an INNO3D iChill Frostbite (water-cooled) at Alternate (Germany), found via PriceSquirrel. A cheaper €7,080 listing, an MSI Ventus 3X OC at Compumsa in Belgium, was out of stock.
Prices excl. shipping unless marked. Euro prices incl. VAT where the seller charges it. Dollar prices are converted from our euro prices without the 21% EU VAT, before US sales tax.
The cheapest graphics card on our list that also runs it (it still needs the rest of a PC): GeForce RTX 3090 24GB, $1,395€1,246, used (median of sold listings). Fast~32 tok/s
A complete machine that runs it: Mac Studio M4 Max 36GB, about $2,116€1,890 (used). Comfortable~8.7 tok/s
If you buy through these links we may earn a commission. It doesn't change our ranking. How shop links work
As an Amazon Associate I earn from qualifying purchases.
Als Amazon-Partner verdiene ich an qualifizierten Verkäufen.
Als Amazon-Partner verdien ik aan in aanmerking komende aankopen.
Nanbeige4.3-33B on other graphics cards
Runs well6
The whole model fits in the graphics memory.
Doesn't fit4
Not enough graphics memory: it needs about 23.3 GB.
- RTX 3060 12GBhas 12 GB
- RTX 4060 Ti 16GBhas 16 GB
- RTX 5080has 16 GB
- RX 9060 XT 16GBhas 16 GB
More about the model itself, what it is good at and its licence: Nanbeige4.3-33B.
The numbers behind this answer
A model comes in several sizes. Q4 is the standard one most people use; the bigger ones answer a little better but need more memory. Each row adds up the model file, the memory a conversation takes, and some working space, and compares the total with what is free on this card.
Swipe the table sideways to see every column.
| Quality | Conversation | Result | Speed | Model file | Conversation memory | Working space | Total needed | Free memory |
|---|---|---|---|---|---|---|---|---|
| Q4 · standard (smallest that keeps quality) | ~12 pages8,192 tokens | Runs | Fast~61 tok/s | 20.8 GBmeasured file | 1.7 GB | 0.8 GB | 23.3 GB | 31.4 GB |
| Q5 · a bit sharper, a bit bigger | ~12 pages8,192 tokens | Runs | Fast~55 tok/s | 23.5 GB | 1.7 GB | 0.8 GB | 26.1 GB | 31.4 GB |
| Q8 · near full quality, almost twice Q4 | ~12 pages8,192 tokens | Doesn't fit | — | 35.4 GB | 1.7 GB | 0.8 GB | 38.0 GB | 31.4 GB |
| FP16 · full quality, over three times Q4 | ~12 pages8,192 tokens | Doesn't fit | — | 66.2 GB | 1.7 GB | 0.8 GB | 68.8 GB | 31.4 GB |
| Q4 · standard, longer conversation | ~49 pages32,768 tokens | Runs | Fast~50 tok/s | 20.8 GBmeasured file | 7.0 GB | 0.8 GB | 28.6 GB | 31.4 GB |
Sources and method
Memory and speed on this page are our estimate, not a measured test. Checked . bartowski Q4_K_M GGUF.
Dense coder and reasoning model, 33.12B parameters, Apache-2.0 on the model-card metadata (the repo had no LICENSE file on 26 Oct 2026). Config: 52 layers, 8 KV heads, head dim 128, context 524,288. The parameter formula is about 19.9 GB at Q4, 35.4 GB at Q8 and 66.2 GB at FP16. Bartowski's GGUF listing on 26 Oct 2026 measured Q4_K_M at 20.79 GB, Q8_0 at 35.20 GB and the two BF16 shards at 66.24 GB together. IQ2_M on the same listing measured 12.19 GB on 27 Oct 2026 (12,187,167,872 bytes). At 8,192 tokens our estimate adds about 1.74 GB of conversation memory and 0.80 GB of working space, so the Q4_K_M total is about 23.3 GB. That fits a 24 GB GPU (an RTX 3090 or RTX 4090 leaves about 23.4 GB usable). A 32 GB Mac leaves about 21.3 GB for the GPU, so the same Q4 total is over that default share; IQ4_XS at 18.64 GB is the file that lands inside it. The IQ2_M total is about 14.7 GB, which fits a 16 GB card and misses a 12 GB card. Q8_0 at 35.20 GB fits 48 GB of VRAM or 64 GB of unified memory. FP16 at 66.24 GB fits 96 GB or 128 GB of unified memory, or two 48 GB GPUs. The Terminal-Bench 80.9 and DeepSWE 65.5 figures are the author's, and are not yet independently verified.
Measured Nanbeige4.3-33B on an RTX 5090 yourself? Send us your llama-bench result.
Questions
- Can I run Nanbeige4.3-33B on an RTX 5090?
- Yes. Nanbeige4.3-33B runs on an RTX 5090, about 61 tokens per second (tok/s).
- How much memory does Nanbeige4.3-33B need?
- About 23.3 GB at the standard quality (Q4) for a conversation of up to about 12 pages of text (8,192 tokens), by our estimate: 20.8 GB for the model file (measured), 1.7 GB to hold the conversation and 0.8 GB of working space.
- How fast is Nanbeige4.3-33B on an RTX 5090?
- Fast: about 61 tokens per second (tok/s). Answers appear faster than you can read them. This is our estimate, not a measured test.
Speeds and memory are our estimate from the hardware specs, not a measured test. Prices are dated and may change. How we calculate