Size
31.2B
3B active — only part of the model works on each answer, but all of it has to fit in memory
GLM-4.7-Flash
Context
200K
the longest conversation the model is built to hold
Lab
Z.ai — GLM
the lab that released it
Released
Jan 19, 2026
Licence: MIT — free for any use, including commercial. Just keep the licence notice.
Worth knowing
Z.ai's lightweight MoE sibling to GLM-4.7, released Jan 19, 2026 — the model Z.ai bills as its 30B-A3B. The checkpoint itself counts 31.2B total parameters by its own safetensors total (47 layers, 64 routed experts with top-4 routing plus one shared expert, per its config), about 3B active per token, a 200K context (config maximum 202,752) and an MIT licence. Public and ungated on Hugging Face, in BF16.
What it needs
Full precision (FP16)
62 GB
the weights, uncompressed
8-bit
31 GB
the weights, compressed to 8-bit
4-bit
16 GB
the weights, compressed to 4-bit
How we work this out: memory = the model's total parameters × bytes per weight — 2 bytes at full precision (FP16), 1 byte at 8-bit, half a byte at 4-bit. Weights only — add room for the conversation memory (KV cache) and the serving software. A GPU counts as running a build when its memory covers the weights plus about 20% headroom. Real figures move with context length and software.
GPUs that can run it
Ranked by how well the model runs, then by price.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 95GB, with room left for the conversation memory.
This is the Max-Q version: it runs at lower power and is slower than the full card.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 80GB, with room left for the conversation memory.
An older generation — no FP8, so the newer speed tricks for serving don't apply.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 96GB, with room left for the conversation memory.
No NVLink scale-out; strong all-round, but the Max-Q version (listed separately) runs slower.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 80GB, with room left for the conversation memory.
No NVLink, and lower power than the SXM — slower when cards must work together.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 80GB, with room left for the conversation memory.
Pays off only when the card stays busy; on bursty work you pay H100 rent for idle silicon.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 94GB, with room left for the conversation memory.
Sold as linked pairs, not as a lone card.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 80GB, with room left for the conversation memory.
One provider's label, published alongside its plainer “H100 PCIe” — check which card you are actually booking.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 141GB, with room left for the conversation memory.
More card than most jobs need; the main trap is paying for it on small work.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 80GB, with room left for the conversation memory.
Pays off only when kept busy — avoid it for bursty development work.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 192GB, with room left for the conversation memory.
Frontier training money. If your job fits on anything cheaper, a B200 is budget burned for a spec sheet.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 96GB, with room left for the conversation memory.
Its processor is ARM, not the usual x86 — some software needs changes to run on it.
Runs GLM-4.7-Flash at full precision — its 62GB of weights fit this card's 288GB, with room left for the conversation memory.
Very costly for small jobs — best value only on big work.
Runs GLM-4.7-Flash compressed — the 8-bit build needs 31GB, which fits this card's 48GB. The compromise: slightly lower answer quality and a shorter conversation memory than the full-precision build.
An older generation — no FP8, so the newer speed tricks for serving don't apply.
Runs GLM-4.7-Flash compressed — the 8-bit build needs 31GB, which fits this card's 48GB. The compromise: slightly lower answer quality and a shorter conversation memory than the full-precision build.
70B at FP16 does not fit, and no flag fixes that. Weak at double-precision maths; no NVLink.
Runs GLM-4.7-Flash compressed — the 8-bit build needs 31GB, which fits this card's 48GB. The compromise: slightly lower answer quality and a shorter conversation memory than the full-precision build.
Weak at double-precision maths — not the card for scientific computing. No NVLink.
Runs GLM-4.7-Flash compressed — the 8-bit build needs 31GB, which fits this card's 40GB. The compromise: slightly lower answer quality and a shorter conversation memory than the full-precision build.
The 40GB fills up quickly on larger models.
Runs GLM-4.7-Flash compressed — the 4-bit build needs 16GB, which fits this card's 24GB. The compromise: slightly lower answer quality and a shorter conversation memory than the full-precision build.
24GB fills up fast on big models. No NVLink, so two cards can't pool memory — and it is a gaming card at heart.
Runs GLM-4.7-Flash compressed — the 4-bit build needs 16GB, which fits this card's 32GB. The compromise: slightly lower answer quality and a shorter conversation memory than the full-precision build.
No NVLink, so two cards can't pool memory. A consumer card — driver and ECC behaviour matter for production serving.
The cheapest verified way to run it
RTX 4090 at RunPod Community — $0.34/hr — runs it compressed — the 4-bit build.
Launch RunPod Community ↗ See the RTX 4090 →
Checked Nov 15, 2026
Select a GPU See these GPUs on the prices board, ranked for this model.