GLM-4.7 Flash 30B-A3B (MoE)
GLM-4.7 Flash 30B-A3B (MoE) needs 19.8 GB of VRAM at Q4_K_M with 8k context, computed from its config.json as checked on 3 November 2026. The smallest card that holds it has 24 GB, and the fastest desktop card that runs it, the RTX 5090, generates at most 721 tokens per second.
- Parameters
- 31.2B, 3B active
- Context
- 202,752 tokens
- Licence
- MIT
- Released
- January 2026
- Config checked
- 3 November 2026
Open GLM-4.7 Flash 30B-A3B (MoE) in the LLM checker
Which card size runs it
19.8 GB against each memory size. A model may use 95% of a card, and only what is left over, or missing, is coloured; tight means above 85%, with little room for a longer conversation.
Other quantisations, context lengths or a Q8 KV cache: the VRAM calculator.
Computed 3 November 2026Which GPU to pick
The smallest card here that loads it at Q4_K_M with 8k of context is the RTX 5090 Laptop GPU (24 GB): 19.8 GB needed, an estimated 360 tokens per second. For room to work, meaning 32,768 tokens of context while staying under 85% of memory, step up to the RTX 5090 (32 GB), which needs 21.1 GB for that and generates at about 721 tokens per second.
At Q8_0, which is near-lossless, the card is the RTX A6000 (48 GB), needing 35.3 GB. On Apple silicon, a Mac mini M4 Pro holds it with 32,768 tokens of context at an estimated 110 tokens per second; prompt processing is slower than on Nvidia, which long documents make obvious. Among unified-memory desktops, the Ryzen AI Max+ 395 holds it at 32,768 tokens in 21.1 GB, at an estimated 103 tokens per second.
Speed on each card
| GPU | Fits | Est. tokens/s | Feels like | Memory | Bandwidth |
|---|---|---|---|---|---|
| H100 SXM | Yes | ≤ 1,348 | faster than you can read | 80 GB | 3,350 GB/s |
| A100 | Yes | ≤ 820 | faster than you can read | 80 GB | 2,039 GB/s |
| RTX 5090 | Yes | ≤ 721 | faster than you can read | 32 GB | 1,792 GB/s |
| RTX PRO 6000 Blackwell | Yes | ≤ 721 | faster than you can read | 96 GB | 1,792 GB/s |
| RTX PRO 5000 Blackwell | Yes | ≤ 541 | faster than you can read | 48 GB | 1,344 GB/s |
| RTX 4090 | Yes | ≤ 406 | faster than you can read | 24 GB | 1,008 GB/s |
| RX 7900 XTX | Yes | ≤ 386 | faster than you can read | 24 GB | 960 GB/s |
| RTX 6000 Ada | Yes | ≤ 386 | faster than you can read | 48 GB | 960 GB/s |
| RTX 3090 | Yes | ≤ 377 | faster than you can read | 24 GB | 936 GB/s |
| RTX 5090 Laptop GPU | Yes | ≤ 360 | faster than you can read | 24 GB | 896 GB/s |
| RTX PRO 4500 Blackwell | Yes | ≤ 360 | faster than you can read | 32 GB | 896 GB/s |
| L40S | Yes | ≤ 348 | faster than you can read | 48 GB | 864 GB/s |
| RTX A6000 | Yes | ≤ 309 | faster than you can read | 48 GB | 768 GB/s |
| RTX PRO 4000 Blackwell | Yes | ≤ 270 | faster than you can read | 24 GB | 672 GB/s |
| Radeon AI PRO R9700 | Yes | ≤ 257 | faster than you can read | 32 GB | 640 GB/s |
| Arc Pro B70 | Yes | ≤ 245 | faster than you can read | 32 GB | 608 GB/s |
| Arc Pro B60 | Yes | ≤ 183 | faster than you can read | 24 GB | 456 GB/s |
| RTX PRO 4000 Blackwell SFF | Yes | ≤ 174 | faster than you can read | 24 GB | 432 GB/s |
| RTX Spark | Yes | ≤ 121 | faster than you can read | 128 GB (102 for the GPU) | 300 GB/s |
| DGX Spark | Yes | ≤ 110 | faster than you can read | 128 GB (126 for the GPU) | 273 GB/s |
| Ryzen AI Max+ 395 | Yes | ≤ 103 | faster than you can read | 128 GB (96 for the GPU) | 256 GB/s |
| RX 7900 XT | No | - | does not fit | 20 GB | 800 GB/s |
| RTX 4060 Ti | No | - | does not fit | 16 GB | 288 GB/s |
| RTX 5060 Ti | No | - | does not fit | 16 GB | 448 GB/s |
| RTX 5080 | No | - | does not fit | 16 GB | 960 GB/s |
| RTX 5070 Ti | No | - | does not fit | 16 GB | 896 GB/s |
| RTX 4080 SUPER | No | - | does not fit | 16 GB | 736 GB/s |
| RX 9070 XT | No | - | does not fit | 16 GB | 640 GB/s |
| RX 9060 XT | No | - | does not fit | 16 GB | 320 GB/s |
| RTX 3060 | No | - | does not fit | 12 GB | 360 GB/s |
| RTX 4070 | No | - | does not fit | 12 GB | 504 GB/s |
| RTX 5070 | No | - | does not fit | 12 GB | 672 GB/s |
| Arc B580 | No | - | does not fit | 12 GB | 456 GB/s |
Single-stream decode, from memory bandwidth at 70% efficiency. Prompt processing and batching not included. These are ceilings, not measurements: GLM-4.7 Flash 30B-A3B (MoE) reads only 1.7 GB of weights per token at Q4, so little that the per-token costs the formula leaves out decide much of the real speed. The 21 figures marked ≤ are where a real runtime falls furthest below the number shown. See the speed estimator for other quantisations and Apple hardware, or every card compared if you are choosing hardware rather than a model.
Computed 3 November 2026Memory by quantisation and context
| Quant | 4,096 ctx | 8,192 ctx | 32,768 ctx |
|---|---|---|---|
| Q8_0 | 35.1 GB · fits 48 GB | 35.3 GB · fits 48 GB | 36.7 GB · fits 48 GB |
| Q5_K_M | 23.8 GB · fits 32 GB | 24 GB · fits 32 GB | 25.3 GB · fits 32 GB |
| Q4_K_M | 19.5 GB · fits 24 GB | 19.8 GB · fits 24 GB | 21.1 GB · fits 24 GB |
Weights at this quant: 18.1 GB at Q4. Every extra 1,000 tokens of context adds about 0.054 GB of KV cache at FP16. Try other settings in the VRAM calculator.
About the model
GLM-4.7 Flash 30B-A3B (MoE) has 31.2B parameters with 3B active per token (mixture of experts). It has 47 layers using multi-head latent attention: instead of keys and values for each of its 20 heads, every layer caches one compressed vector of 576 values per token. Z.ai's 30B-A3B mixture-of-experts model, January 2026, the lightweight member of the GLM-4.7 line. MIT-licensed, text only, with latent attention that keeps its cache small.
Text only 202,752-token window MIT January 2026 3B of 31.2B active
Reasoning. Supports thinking, including a preserved-thinking mode for multi-turn agent work.
A runtime that does not implement the latent cache stores full keys and values and needs several times more context memory.
Native context window: 202,752 tokens. Licence: MIT, a permissive licence. Weights published in January 2026.
Good at
- Agentic and tool-using tasks on a 24 GB card, which is how Z.ai positions it.
- The most permissive licence among the 30B-class mixture-of-experts models here.
Watch out for
- The model card names vLLM and SGLang as its runtimes. Before relying on the context figures here, check that yours implements the latent cache; one that does not stores full keys and values for 20 heads and needs several times more.
- Text only.
Similar models
The models that need about the same memory as GLM-4.7 Flash 30B-A3B (MoE) at Q4_K_M and 8k, which makes them the real alternatives on whatever card you have. The context column is what each extra 1,000 tokens costs, and it is where models of the same size differ most.
| Model | Size | Needs | Per 1k context | Window | Released |
|---|---|---|---|---|---|
| GLM-4.7 Flash 30B-A3B (MoE) | 31.2B · 3B active | 19.8 GB | 0.054 GB | 198k | 2026-01 |
| Qwen3 30B-A3B (MoE) | 30.5B · 3.3B active | 19.7 GB | 0.098 GB | 40k | 2025-04 |
| Nemotron 3.5 Lightning 30B-A3B (MoE) | 31.6B · 3B active | 19.7 GB | 0.006 GB | 256k | 2026-08 |
| Granite 4.2 30B | 29.3B | 20.3 GB | 0.262 GB | 128k | 2026-08 |
| Gemma 4 31B | 31.3B | 20.9 GB | 0.082 GB | 256k | 2026-03 |
See every model compared, which models fit on 8 to 96 GB, or the cheapest hour of every cloud card.
Questions
- Can I run GLM-4.7 Flash 30B-A3B (MoE) on a 24 GB card?
- Yes, at Q4_K_M and 8k context it needs about 19.8 GB, which fits a 24 GB card with room.
- How much VRAM does GLM-4.7 Flash 30B-A3B (MoE) need at Q8?
- About 35.3 GB at 8k context, or 36.7 GB at 32k. Q8 is near-lossless; use it when it fits.
- What is the context length of GLM-4.7 Flash 30B-A3B (MoE)?
- 202,752 tokens natively. Holding all of it at Q4_K_M takes about 30.3 GB, of which 11 GB is context.
- Why does GLM-4.7 Flash 30B-A3B (MoE) need so little memory for long context?
- Because most of its layers do not keep a cache that grows. It has 47 layers using multi-head latent attention: instead of keys and values for each of its 20 heads, every layer caches one compressed vector of 576 values per token. At 32k tokens the context costs about 1.8 GB; if all 47 layers kept a conventional full-length cache, the same conversation would cost about 31.5 GB.
- How fast is GLM-4.7 Flash 30B-A3B (MoE) on an RTX 4090?
- At most about 406 tokens per second at Q4, single stream, which is faster than you can read. That is a ceiling worked out from the card's memory bandwidth, not a measurement, and at a figure this high real runtimes land well below it, because the costs the formula leaves out take a growing share of each token.
- Architecture values from config.json in zai-org/GLM-4.7-Flash on Hugging Face, and its model card, checked 3 November 2026. Every figure on this page is computed from those values; check them before buying hardware for this model.
- Estimates assume a single conversation on a card that is otherwise free. A desktop on the same GPU takes 0.5 to 2 GB.