GLM-4.7-Flash-AWQ
Then, create a fresh Python environment (e.g. python3.12 venv) and run: bash pip install -U vllm --pre --index-url https://pypi.org/simple --extra-index-url https://wheels.vllm.ai/nightly pip install git+https://github.com/huggingface/transformers.git
Params
31.2 B
Context
202,752
Downloads 30d
447 K
Likes
12
Download history
daily snapshots · 83 days
▲ 270 K in the last 30 days (152.5%)
463 K177 K
Sep 25Oct 5Oct 15Oct 24
463 K171 K
Aug 3Aug 30Sep 27Oct 24
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| model.safetensors | i32 | 19.8 GB | 26.9 GB | ❌ Won’t fit |
| model.safetensors (bf16, full) | bf16 + 198K ctx | 19.8 GB | 138.1 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ curl -s https://aimodelscomparison.com/api/v1/models/glm-4-7-flash-awq
{
"hf_id": "QuantTrio/GLM-4.7-Flash-AWQ",
"params_b": 31.22,
"context_length": 202752,
"license": { "id": "mit", "commercial": "yes" },
"downloads_30d": 447184,
"vram_estimates": [
{ "quant": "i32", "gb": 26.9 }
],
"updated_at": "2026-08-03T01:00:38Z"
}
Specifications
- Architecture
- Glm4MoeLiteForCausalLM
- Parameters
- 31.2 B
- Tensor type
- I32
- Context length
- 202,752
- Vocabulary
- 154,880
- Layers / heads
- 47 / 20
- Licence
- mit
- First seen on the Hub
- 2026-01-21
- Base model
- GLM-4.7-Flash
- Training datasets
- undisclosed
- Added to our catalog
- 2026-08-03
Family
Base model and the most-downloaded derivatives in the catalog.
Compare with any text-generation model