Cactus-Compute / text-generation Open weights updated 4 days ago

needle3

Name: needle3
Author: Cactus-Compute

An on-device model for tool calling, structured extraction and text embeddings. See the Needle website for model details, benchmarks and the browser demo.

Params

—

Context

8,192

Downloads 30d

136 K

Likes

406

Commercial use: allowed apache-2.0 Not gated SAFETENSORS View on Hugging Face ↗

Download history

daily snapshots · 24 days

161 K136 K

Oct 18Oct 26Nov 3Nov 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	fp16	0.2 GB	0.8 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/needle3
{
  "hf_id": "Cactus-Compute/needle3",
  "params_b": null,
  "context_length": 8192,
  "license": { "id": "apache-2.0", "commercial": "yes" },
  "downloads_30d": 136348,
  "vram_estimates": [
    { "quant": "fp16", "gb": 0.8 }
  ],
  "updated_at": "2026-11-08T01:00:23Z"
}

est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture: NeedleForToolCalling
Context length: 8,192
Vocabulary: 8,192
Layers / heads: 20 / 12
Licence: apache-2.0
First seen on the Hub: 2026-09-16
Training datasets: undisclosed
Added to our catalog: 2026-10-18

Compare with any text-generation model