Fact sheet
Parameters
Context length
Architecture
Shape
Weights stored as
Task
Languages
Built on
Released
Last updated
Downloads (last 30 days)
Licence and access
Licence
Access
Security
Hugging Face security scan
Weight files by format
Facthory’s checks
Safetensors or GGUF files beside pickle-based ones.
No file flagged.
The card declares its licence.
The model card is not public.
GPU calculator
How much memory this model needs to serve at a quantization, context and number of concurrent sequences, which GPUs hold it, and how fast it can generate. Estimates from the model’s published size and shape, with the assumptions below.
Weights
Context per sequence: 8K tokens
Concurrent sequences: 1
KV cache
Memory
18 GB
Weights: 16 GB
KV cache: –
Runtime: 1.6 GB
The model’s shape is not published, so the KV cache is not included.
| GPU | Memory | GPUs needed | Per sequence | All sequences |
|---|---|---|---|---|
| NVIDIA L4 | 24 GB | 1 × | 11 tokens/s | 11 tokens/s |
| NVIDIA L40S | 48 GB | 1 × | 32 tokens/s | 32 tokens/s |
| NVIDIA RTX 6000 Ada | 48 GB | 1 × | 36 tokens/s | 36 tokens/s |
| NVIDIA RTX PRO 6000 Blackwell | 96 GB | 1 × | 67 tokens/s | 67 tokens/s |
| NVIDIA A100 80GB SXM | 80 GB | 1 × | 76 tokens/s | 76 tokens/s |
| NVIDIA H100 SXM | 80 GB | 1 × | 125 tokens/s | 125 tokens/s |
| NVIDIA H200 SXM | 141 GB | 1 × | 179 tokens/s | 179 tokens/s |
| AMD Instinct MI300X | 192 GB | 1 × | 198 tokens/s | 198 tokens/s |
| AMD Instinct MI325X | 256 GB | 1 × | 224 tokens/s | 224 tokens/s |
Assumptions
- Runtime memory (activations, buffers) is 10% of weights and KV cache, plus 1 GB per GPU for the framework.
- A GPU is filled to at most 90% of its memory; tensor parallelism uses 1, 2, 4 or 8 GPUs in one server.
- Generation is limited by memory bandwidth at 60% of the vendor’s peak: each step reads the weights once and every sequence’s KV cache. A mixture-of-experts model reads only its active experts, so it runs faster than shown.
- GPU memory and bandwidth are the vendors’ published figures.
Profile
Kind
Lineage
In and out
Written on 12 Nov 2026 by Workers AI (@cf/zai-org/glm-4.7-flash) from the hub’s metadata: a reading of the publisher’s text, not checked by Facthory.
Published benchmarks
The model card publishes no benchmark results.
EU AI Act
What the EU AI Act asks of a general-purpose AI model depends on who places it on the EU market and how. These facts bear on it:
- Licence: llama3
- Access: Weights download after the publisher approves the request
- Parameters: 8B
General-purpose AI model (Article 3(63))
Indicated. Epoch AI estimates its training compute at 7.2 × 10²³ FLOP (as Meta-Llama-3-8B), above the 10²³ FLOP of the Commission’s indicative criterion, and it generates text, audio, images or video.
Systemic risk (Article 51)
Not presumed. Its training compute, 7.2 × 10²³ FLOP by Epoch AI’s estimate (as Meta-Llama-3-8B), is below the 10²⁵ FLOP of Article 51(2).
Open-source exemption (Article 53(2))
Does not apply. Its licence (llama3) is not a free and open-source licence, one that lets anyone use, study, change and share the model.
General-Purpose AI Code of Practice
Meta is not on the European Commission’s list of signatories. The Code is voluntary.
What the card and the provider document
The model card is not public.
The model card is not public.
