zai-org / text-generation Open weights Also via API updated 1 year ago

GLM-4.5

πŸ‘‹ Join our Discord community. πŸ“– Check out the GLM-4.5 technical blog, technical report, and Zhipu AI technical documentation. πŸ“ Use GLM-4.5 API services on Z.ai API Platform (Global) or Zhipu AI Open Platform (Mainland China). πŸ‘‰ One click to GLM-4.5.

Params
358.3 B
Context
131,072
Downloads 30d
114 K
Likes
1,419
Commercial use: allowed mit Not gated SAFETENSORS 2 languages View on Hugging Face β†—

Download history

tracking started β€” chart appears after 7 days of snapshots (5 recorded)
114 K downloads in the last 30 days

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

FileQuantSizeEst. VRAMVerdict on RTX 4090 Β· 24 GB
model.safetensors bf16 716.7 GB 842.6 GB ❌ Won’t fit
model.safetensors (bf16, full) bf16 + 128K ctx 716.7 GB 1,648.9 GB ❌ Won’t fit
Estimate: file size Γ— 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark β€” how we calculate this.

Run it

copy-paste, exact tags checked against the Hub
~ Β· curl Β· api/v1
$ curl -s https://aimodelscomparison.com/api/v1/models/glm-4-5
{
  "hf_id": "zai-org/GLM-4.5",
  "params_b": 358.34,
  "context_length": 131072,
  "license": { "id": "mit", "commercial": "yes" },
  "downloads_30d": 113929,
  "vram_estimates": [
    { "quant": "bf16", "gb": 842.6 }
  ],
  "updated_at": "2026-10-19T06:00:46Z"
}
est. VRAM β€”on RTX 4090 Β· 24 GBJSON API β†’

Use via API

via OpenRouter Β· USD per 1M tokens Β· synced 2026-10-24
$0.60
input / 1M
$2.20
output / 1M
$0.11
cached input / 1M
$2.80
1M in + 1M out
Context window
131,072 tokens
Max output
98,304 tokens
Providers
1
Cheapest provider
Z.AI Β· $0.60 / $2.20
Best throughput
β€”
AA Intelligence / Coding / Agentic
12.8 / β€” / β€”
βœ“ Tool calling βœ“ Reasoning βœ“ Structured outputs βœ— Image input βœ— File input βœ— Audio input βœ“ Prompt caching

Specifications

Architecture
Glm4MoeForCausalLM
Parameters
358.3 B
Tensor type
BF16
Context length
131,072
Vocabulary
151,552
Layers / heads
92 / 96
Licence
mit
First seen on the Hub
2025-07-20
Training datasets
undisclosed
Added to our catalog
2026-10-20
Compare with any text-generation model