LeaderboardModel1 / text-generation updated 3 months ago

zeta-2.1-autoround-W4A16

This model is a int4 weight-only quantization with groupsize 128 and symmetric quantization of zed-industries/zeta-2.1 generated by AutoRound. Please follow the license of the original model.

Params
2.2 B
Context
32,768
Downloads 30d
427 K
Likes
0
Licence: unknown Not gated SAFETENSORS View on Hugging Face ↗

Download history

daily snapshots · 14 days
427 K197 K
Sep 7Sep 11Sep 16Sep 20

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 · 24 GB
model.safetensors bf16 6.2 GB 7.6 GB ✅ Runs comfortably
model.safetensors (bf16, full) bf16 + 32K ctx 6.2 GB 8.6 GB ✅ Runs comfortably
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 · api/v1
$ curl -s https://aimodelscomparison.com/api/v1/models/zeta-2-1-autoround-w4a16
{
  "hf_id": "LeaderboardModel1/zeta-2.1-autoround-W4A16",
  "params_b": 2.20,
  "context_length": 32768,
  "license": { "id": "", "commercial": "unknown" },
  "downloads_30d": 427115,
  "vram_estimates": [
    { "quant": "bf16", "gb": 7.6 }
  ],
  "updated_at": "2026-09-07T01:00:23Z"
}
est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture
LlamaForCausalLM
Parameters
2.2 B
Tensor type
BF16
Context length
32,768
Vocabulary
155,136
Layers / heads
32 / 32
First seen on the Hub
2026-05-21
Training datasets
undisclosed
Added to our catalog
2026-09-07
Compare with any text-generation model