How much memory does GLM-5.3-Flash need at 4-bit?

Unsloth's UD-Q4_K_XL GGUF is 199.71 GB, so our estimate is about 200.7 GB with 8,192 tokens of context and working space. That is just over the 192 GB usable on a 256 GB Mac. Smaller 4-bit files exist: UD-IQ4_XS is 156.82 GB and ggml-org's Q4_K is 181.37 GB.

Can GLM-5.3-Flash run on a 128 GB Mac or Strix Halo mini PC?

Only at 1-bit. Unsloth's UD-IQ1_S file is 93.09 GB, about 94.1 GB with 8,192 tokens of context, inside the 96 GB we count as usable on 128 GB Macs and Strix Halo boxes, with under 2 GB to spare. The next file up (UD-IQ1_M, 97.58 GB) does not fit. Unsloth says 3-bit works on 128 GB devices; that relies on more memory than our rule allows and we have not tested it.

How much context memory does GLM-5.3-Flash use?

Very little. Only 11 of its 45 layers use attention that grows with context (sparse attention with a compressed 512-wide cache); the other 34 are linear-attention layers with a fixed-size state. Our estimate is 0.18 GB at 8,192 tokens, excluding that fixed state and the indexer cache.

Does llama.cpp or MLX support GLM-5.3-Flash?

Yes. Mainline llama.cpp added GLM-5.3-Flash support in pull request 27773, merged on 30 September 2026, and ggml-org, Unsloth and others publish GGUF files. MLX builds exist too, for example mlx-community/GLM-5.3-Flash-4bit at 203.99 GB and pipenetwork's 4-bit build at 177.55 GB.

Is GLM-5.3-Flash free for commercial use?

Yes. Z.ai publishes the weights under the MIT licence, which allows commercial use. Read the licence on the model page before you ship.

Can I run GLM-5.3-Flash locally?

Key facts

Yes, on a 256 GB Mac Studio M5 Ultra at about 4-bit: Unsloth's UD-IQ4_XS file needs about 157.8 GB with 8,192 tokens of context, inside the 192 GB we count as usable. The standard UD-Q4_K_XL file (about 200.7 GB) is just too big, and 128 GB Macs and Strix Halo mini PCs only fit the 1-bit file.

About 200.7 GB at UD-Q4_K_XL, our estimate. Checked 29 Oct 2026. Unsloth UD-Q4_K_XL GGUF.

GLM-5.3-Flash has about 320B parameters with 18B active per token, MIT-licensed weights, image input and a 1M-token context window. Only 11 of its 45 layers keep a growing context cache, so long context costs little memory. If you want a model that fits a 24 GB card, see what fits 24 GB, or work out your own setup in the VRAM calculator. A 128 GB unified-memory box is on what fits 128 GB. Compare GLM-5.3, MiniMax M3.1 Flash and DeepSeek V4.1 Flash. Machines: Mac Studio M5 Ultra 256GB, Mac Studio M5 Max 128GB and Strix Halo 128GB. Every model is on the model list.

None of the 43 graphics cards, PCs and Macs on our list holds the whole model at the standard quality in its graphics memory.

The smallest download, a file of 93.1 GB, does fit on one: Mac Studio M4 Max 128GB, about $3,683€3,290 (used). It answers a little less well than the standard version.

Our price is for a used Mac. These shops mostly sell new ones, so their price can be higher.

EU shops · Shown to visitors in Europe.

US shops · Shown to visitors outside Europe.

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.

We never pass off an estimate as a benchmark. Commissions never change which hardware we recommend. How we calculate

Find hardware for GLM-5.3-Flash About this model

LIVE IN BROWSER · $0 · NO LOGIN

Test comparable open models (Recursive Flint 40B vs Gemma 4 31B) free in your browser

Open full 8-model playground ↗

Click a starter prompt below or type your own question to watch Recursive Flint stream side by side against Gemma 4 31B right on this page:

Which graphics cards run it

Seven popular graphics cards and two PCs with two cards each, at the standard quality (Q4). Click one for the full answer: memory, speed and price.

Doesn't fit10

Not enough graphics memory: it needs about 200.7 GB.

RTX 3060 12GBhas 12 GB
RTX 4060 Ti 16GBhas 16 GB
RTX 3090has 24 GB
RTX 4090has 24 GB
RTX 5080has 16 GB
RX 9060 XT 16GBhas 16 GB
RTX 5090has 32 GB
RX 7900 XTXhas 24 GB
Dual RTX 3090has 48 GB
Dual RTX 4090has 48 GB

Every model on every card

GLM-5.3-Flash needs 200.7 GB at Q4. When a new open model would fit 128 GB, or the cards and Macs that run it change price, The Fit List tells you, with the memory it needs and dated EU/US prices. Plus a free GPU buying cheat sheet (PDF) now.

Which Macs and mini PCs run it

Macs and mini PCs from the picker, at the standard quality (Q4). A Mac shares its memory with the rest of the computer. Each name opens that machine's page.

Doesn't fit14

Not enough shared memory: it needs about 200.7 GB.

Mac mini M4 16GBhas 16 GB
Mac mini M6 32GBhas 32 GB
Mac Studio M5 Max 36GBhas 36 GB
Mac mini M4 Pro 64GBhas 64 GB
Mac mini M5 Pro 64GBhas 64 GB
GMKtec EVO-X2 (Ryzen AI Max+ 395) 64GBhas 64 GB
NVIDIA DGX Spark 64GBhas 64 GB
Mac Studio M5 Ultra 96GBhas 96 GB
GMKtec EVO-X2 (Ryzen AI Max+ 395) 128GBhas 128 GB
NVIDIA DGX Spark 128GBhas 128 GB
GEEKOM A9 Mega (Ryzen AI Max+ 395) 128GBhas 128 GB
Beelink GTR9 Pro (Ryzen AI Max+ 395) 128GBhas 128 GB
Mac Studio M5 Max 128GBhas 128 GB
Mac Studio M5 Ultra 256GBhas 256 GB

Cheapest hardware for each size

From all 43 graphics cards, PCs and Macs on our list. "Runs slowly" means part of the model runs from normal memory: cheaper, but much slower. A smaller download needs less memory and answers a little less well.

Standard quality (UD-Q4_K_XL)No machine on our list runs this size.
Standard quality (UD-Q4_K_XL), runs slowlyNo machine on our list runs this size.
Smallest download (UD-IQ1_S)Mac Studio M4 Max 128GB, about $3,683€3,290 (used)
Smallest download (UD-IQ1_S), runs slowlyDual RTX 3090, 64 GB RAM, about $3,808€3,592 (with used graphics cards)

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.

Prices moved: see Building a local LLM rig in the RAMageddon.

About GLM-5.3-Flash

321.32B total, 18B active. Parameter count is an estimate. More parameters usually means a smarter model that needs more memory.

Licence:: MIT source
Longest conversation:: about 1573 pages of text (1,048,576 tokens)
Input:: text+image
Released:: 26 Aug 2026 source

Sources and method

Z.ai's open-weight multimodal mixture of experts (MIT licence), the smaller sibling of GLM-5.3. The model card says 320B total parameters with 18B active per token. Hugging Face safetensors metadata lists 321,323,031,390 parameters (F8_E4M3 314.40B, BF16 6.93B, F32 295,518), including the vision encoder and 1 MTP layer, stored here as 321.32B, so our count is an estimate. Config (Glm5NextForConditionalGeneration, text_config glm5_next_text): 45 layers (first 3 dense), hidden size 4096, 288 routed experts plus 1 shared, 8 routed per token, 1 MTP layer, max_position_embeddings 1,048,576. Attention is hybrid: 34 linear-attention layers (KDA delta-rule linear attention, 64 heads x 128) and 11 DeepSeek-style sparse-attention layers (MLA, kv_lora_rank 512, qk_rope_head_dim 0, sparse indexer 32 heads x 128, top-k 2048), every 4th layer from layer 3. Our estimate counts conversation memory only for the 11 sparse-attention layers (latent width 512, K+V formula), 0.18 GB at 8,192 tokens. Linear-attention layers are stored with a sliding window of 0 so our estimate gives them no conversation memory; their fixed recurrent state and the indexer key cache are not counted. Q4 uses the measured Unsloth UD-Q4_K_XL file, 199.71 GB (199,707,321,472 bytes). Q5, Q8 and FP16 are our bytes-per-parameter estimates (0.71 / 1.07 / 2.0 x 321.32B). Unsloth GGUF sizes, shards summed from the Hugging Face tree, decimal GB, checked 29 Oct 2026 (https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF): UD-IQ1_S 93.09, UD-IQ1_M 97.58, UD-IQ2_XXS 101.84, UD-Q2_K_XL 108.72, UD-IQ3_XXS 120.37, UD-Q3_K_XL 147.54, UD-IQ4_XS 156.82, UD-Q4_K_XL 199.71, UD-Q5_K_XL 240.31, UD-Q6_K_XL 291.83, Q8_0 340.98, BF16 641.64; vision mmproj-F16 1.13 GB extra. ggml-org/GLM-5.3-Flash-GGUF: Q4_K 181.37. antirez/glm-5.3-flash-gguf: Q2 96.51, Q4_K 190.88. MLX: mlx-community/GLM-5.3-Flash-4bit 203.99 GB; pipenetwork/GLM-5.3-Flash-MLX-4bit 177.55 GB. NVIDIA NVFP4 checkpoint 204.44 GB. Official FP8 weights are 328.34 GB. Pruned derivatives are not GLM-5.3-Flash and are not scored. Mainline llama.cpp added GLM-5.3-Flash support in pull request 27773, merged 30 September 2026 (https://github.com/ggml-org/llama.cpp/pull/27773). Unsloth says the 1-bit quant works on 100 GB RAM and 3-bit on 128 GB devices (https://unsloth.ai/docs/models/glm-5.3-flash); that is a vendor claim, not tested by us. On our rule a 256 GB Mac Studio M5 Ultra (192 GB usable) holds UD-IQ4_XS (157.80 GB with context) but not UD-Q4_K_XL (200.69 GB); a 128 GB box (96 GB usable) holds only UD-IQ1_S (94.07 GB). Released 26 August 2026 (Artificial Analysis); Hugging Face repo created 25 August 2026. Artificial Analysis Intelligence Index v5.0: not scored yet (checked 29 Oct 2026). Thinking modes low / high / max. Ollama lists only a cloud tag (glm-5.3-flash:cloud), checked 29 Oct 2026.

Ran GLM-5.3-Flash on your own hardware? Send us your llama-bench result.

Parameter source · Architecture source · Model page

Memory at each quality level

Each download of GLM-5.3-Flash is a different size. The total adds the file, the memory a conversation of about 12 pages of text (8,192 tokens) takes, and 0.8 GB of working space. A measured file size is labelled; the rest is our estimate. The last column is the cheapest machine that holds the total.

Swipe the table sideways to see every column, up to the cheapest machine.

Memory per quant for GLM-5.3-Flash
Quality (file)	File size	Conversation	Working space	Total	Cheapest machine that holds it
BF16	641.64 GBmeasured GGUF	0.2 GB	0.8 GB	642.6 GB	No machine on our list runs this size.
Q8_0	340.98 GBmeasured GGUF	0.2 GB	0.8 GB	342.0 GB	No machine on our list runs this size.
UD-Q6_K_XL	291.83 GBmeasured GGUF	0.2 GB	0.8 GB	292.8 GB	No machine on our list runs this size.
UD-Q5_K_XL	240.31 GBmeasured GGUF	0.2 GB	0.8 GB	241.3 GB	No machine on our list runs this size.
UD-Q4_K_XL	199.71 GBmeasured GGUF	0.2 GB	0.8 GB	200.7 GB	No machine on our list runs this size.
Q4_K	190.88 GBmeasured GGUF	0.2 GB	0.8 GB	191.9 GB	Mac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
Q4_K	181.37 GBmeasured GGUF	0.2 GB	0.8 GB	182.4 GB	Mac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-IQ4_XS	156.82 GBmeasured GGUF	0.2 GB	0.8 GB	157.8 GB	Mac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-Q3_K_XL	147.54 GBmeasured GGUF	0.2 GB	0.8 GB	148.5 GB	Mac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-IQ3_XXS	120.37 GBmeasured GGUF	0.2 GB	0.8 GB	121.4 GB	Mac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-Q2_K_XL	108.72 GBmeasured GGUF	0.2 GB	0.8 GB	109.7 GB	Mac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-IQ2_XXS	101.84 GBmeasured GGUF	0.2 GB	0.8 GB	102.8 GB	Mac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
Q2	96.51 GBmeasured GGUF	0.2 GB	0.8 GB	97.5 GB	Mac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-IQ1_M	97.58 GBmeasured GGUF	0.2 GB	0.8 GB	98.6 GB	Mac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-IQ1_S	93.09 GBmeasured GGUF	0.2 GB	0.8 GB	94.1 GB	Mac Studio M4 Max 128GB, $3,683€3,290 (used), price unverified

speed: not estimated for MoE yet.

Speeds and memory are our estimate from the hardware specs, not a measured test. Prices are dated and may change. How we calculate

Questions

Can I run GLM-5.3-Flash locally?: Yes, on a large unified-memory machine, at a lower quant. GLM-5.3-Flash has about 320B parameters (18B active). With 8,192 tokens of context, Unsloth's UD-IQ4_XS file needs about 157.8 GB, which fits the 192 GB usable on a 256 GB Mac Studio M5 Ultra. No graphics card or dual-GPU build on our list holds it.
How much memory does GLM-5.3-Flash need at 4-bit?: Unsloth's UD-Q4_K_XL GGUF is 199.71 GB, so our estimate is about 200.7 GB with 8,192 tokens of context and working space. That is just over the 192 GB usable on a 256 GB Mac. Smaller 4-bit files exist: UD-IQ4_XS is 156.82 GB and ggml-org's Q4_K is 181.37 GB.
Can GLM-5.3-Flash run on a 128 GB Mac or Strix Halo mini PC?: Only at 1-bit. Unsloth's UD-IQ1_S file is 93.09 GB, about 94.1 GB with 8,192 tokens of context, inside the 96 GB we count as usable on 128 GB Macs and Strix Halo boxes, with under 2 GB to spare. The next file up (UD-IQ1_M, 97.58 GB) does not fit. Unsloth says 3-bit works on 128 GB devices; that relies on more memory than our rule allows and we have not tested it.
How much context memory does GLM-5.3-Flash use?: Very little. Only 11 of its 45 layers use attention that grows with context (sparse attention with a compressed 512-wide cache); the other 34 are linear-attention layers with a fixed-size state. Our estimate is 0.18 GB at 8,192 tokens, excluding that fixed state and the indexer cache.
Does llama.cpp or MLX support GLM-5.3-Flash?: Yes. Mainline llama.cpp added GLM-5.3-Flash support in pull request 27773, merged on 30 September 2026, and ggml-org, Unsloth and others publish GGUF files. MLX builds exist too, for example mlx-community/GLM-5.3-Flash-4bit at 203.99 GB and pipenetwork's 4-bit build at 177.55 GB.
Is GLM-5.3-Flash free for commercial use?: Yes. Z.ai publishes the weights under the MIT licence, which allows commercial use. Read the licence on the model page before you ship.