RedHatAI / text-generation updated 10 months ago

Apertus-70B-Instruct-2509-quantized.w4a16

Name: Apertus-70B-Instruct-2509-quantized.w4a16
Author: RedHatAI

- Model Architecture: ApertusForCausalLM - Input: Text - Output: Text - Model Optimizations: - Weight quantization: INT4 - Release Date: 9/22/2025 - Version: 1.0 - Model Developers: Red Hat

Params

11.3 B

Context

65,536

Downloads 30d

338 K

Likes

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

Download history

daily snapshots · 10 days

338 K226 K

Jul 28Jul 31Aug 3Aug 6

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	39.9 GB	46.0 GB	❌ Won’t fit
model.safetensors (bf16, full)	bf16 + 64K ctx	39.9 GB	57.9 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 · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/apertus-70b-instruct-2509-quantized-w4a16
{
  "hf_id": "RedHatAI/Apertus-70B-Instruct-2509-quantized.w4a16",
  "params_b": 11.31,
  "context_length": 65536,
  "license": { "id": "apache-2.0", "commercial": "yes" },
  "downloads_30d": 337908,
  "vram_estimates": [
    { "quant": "i32", "gb": 46.0 }
  ],
  "updated_at": "2026-07-28T18:07:39Z"
}

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

Specifications

Architecture: ApertusForCausalLM
Parameters: 11.3 B
Tensor type: I32
Context length: 65,536
Vocabulary: 131,072
Layers / heads: 80 / 64
Licence: apache-2.0
First seen on the Hub: 2025-09-21
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Compare with

Qwen3-8B 16.3 M Qwen2.5-7B-Instruct 12.1 M Qwen3.6-35B-A3B-NVFP4 11.3 M gpt-oss-20b 8.6 M Llama-3.1-8B-Instruct 8.0 M granite-4.1-8b 4.2 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB