cyankiwi / text-generation updated 1 month ago

MiniCPM-SALA-AWQ-8bit

> [!NOTE] > ### 🏆 2026 Sparse Operator Acceleration & Race (SOAR) is Now Live! > > "The MiniCPM-SALA architecture is just the beginning. Realizing its full potential requires deep system-level synergy and cross-layer compilation optimization." > > In collaboration with SGLang and NVIDIA, OpenBMB invites global geeks to push the boundaries of 9B-scale, 1M-token inference on NVIDIA 6000D. > > 💰 Prize Pool: >$100,000...

Params
3.1 B
Context
524,288
Downloads 30d
616 K
Likes
0
Commercial use: allowed apache-2.0 Not gated SAFETENSORS 2 languages View on Hugging Face ↗

Download history

daily snapshots · 41 days
▲ 186 K in the last 30 days (43.3%)
619 K166 K
Aug 12Aug 25Sep 8Sep 21

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 i32 10.6 GB 12.7 GB ✅ Runs comfortably
model.safetensors (bf16, full) bf16 + 512K ctx 10.6 GB 42.0 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/minicpm-sala-awq-8bit
{
  "hf_id": "cyankiwi/MiniCPM-SALA-AWQ-8bit",
  "params_b": 3.10,
  "context_length": 524288,
  "license": { "id": "apache-2.0", "commercial": "yes" },
  "downloads_30d": 615964,
  "vram_estimates": [
    { "quant": "i32", "gb": 12.7 }
  ],
  "updated_at": "2026-08-12T01:00:35Z"
}
est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture
MiniCPMSALAForCausalLM
Parameters
3.1 B
Tensor type
I32
Context length
524,288
Vocabulary
73,448
Layers / heads
32 / 32
Licence
apache-2.0
First seen on the Hub
2026-02-15
Training datasets
undisclosed
Added to our catalog
2026-08-12
Compare with any text-generation model