GLM-5.3
GLM-5.3 uses the same base model as GLM-5.2 — every gain comes from post-training. Compared with GLM-5.2, it is much better at complex coding and long-horizon tasks:
Params
753.3 B
Context
1,048,576
Downloads 30d
2.0 M
Likes
2,401
Download history
daily snapshots · 67 days
▲ 347 K in the last 30 days (20.8%)
2.5 M1.7 M
Oct 12Oct 22Nov 1Nov 10
2.5 M304 K
Sep 5Sep 27Oct 19Nov 10
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 | f8_e4m3 | 755.6 GB | 944.7 GB | ❌ Won’t fit |
| model.safetensors (bf16, full) | bf16 + 1024K ctx | 755.6 GB | 15,295.6 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-5-3
{
"hf_id": "zai-org/GLM-5.3",
"params_b": 753.33,
"context_length": 1048576,
"license": { "id": "other", "commercial": "unknown" },
"downloads_30d": 2016722,
"vram_estimates": [
{ "quant": "f8_e4m3", "gb": 944.7 }
],
"updated_at": "2026-10-15T01:00:04Z"
}
Use via API
via OpenRouter · USD per 1M tokens · synced 2026-11-10$0.07
input / 1M
$1.40
output / 1M
$0.069
cached input / 1M
$1.47
1M in + 1M out
:batch variant
$0.45 in · $2.00 out
- Context window
- 1,048,576 tokens
- Max output
- 943,718 tokens
- Providers
- 36
- Cheapest provider
- Wafer · $0.07 / $1.40
- Best throughput
- —
- AA Intelligence / Coding / Agentic
- 31.7 / 74.8 / 53.1
✓ Tool calling
✓ Reasoning
✓ Structured outputs
✗ Image input
✗ File input
✗ Audio input
✓ Prompt caching
Specifications
- Architecture
- GlmMoeDsaForCausalLM
- Parameters
- 753.3 B
- Tensor type
- F8_E4M3
- Context length
- 1,048,576
- Vocabulary
- 154,880
- Layers / heads
- 78 / 64
- Licence
- other
- First seen on the Hub
- 2026-08-25
- Training datasets
- undisclosed
- Added to our catalog
- 2026-09-05
Family
Base model and the most-downloaded derivatives in the catalog.
Compare with any text-generation model