Hy-Memory-30B-A3B-GGUF
Community GGUF quantizations of Tencent Hy-Memory-30B-A3B, a 30B-parameter MoE memory-writing model with approximately 3B active parameters. Original model by Tencent Hunyuan; conversion and quantization by rapatsky. No additional training was performed.
Params
30.0 B
Context
—
Downloads 30d
134 K
Likes
2
Download history
tracking started — chart appears after 7 days of snapshots (1 recorded)134 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.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| Hy-Memory-30B-A3B-Q4_K_M.gguf | Q4_K_M | 18.2 GB | 25.1 GB | ❌ Won’t fit |
| Hy-Memory-30B-A3B-Q5_K_M.gguf | Q5_K_M | 21.4 GB | 28.5 GB | ❌ Won’t fit |
| Hy-Memory-30B-A3B-Q6_K.gguf | Q6_K | 24.7 GB | 32.2 GB | ❌ Won’t fit |
| Hy-Memory-30B-A3B-Q8_0.gguf | Q8_0 | 32.0 GB | 40.2 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$ ollama run hy-memory-30b-a3b-gguf # pin the quantization explicitly $ ollama run hy-memory-30b-a3b-gguf-q4_k_m
$ huggingface-cli download rapatsky/Hy-Memory-30B-A3B-GGUF-GGUF \
Hy-Memory-30B-A3B-Q4_K_M.gguf --local-dir .
$ llama-cli -m Hy-Memory-30B-A3B-Q4_K_M.gguf \
-c 8192 -ngl 99 -t 8 --color
$ curl -s https://aimodelscomparison.com/api/v1/models/hy-memory-30b-a3b-gguf
{
"hf_id": "rapatsky/Hy-Memory-30B-A3B-GGUF",
"params_b": 30.00,
"context_length": null,
"license": { "id": "apache-2.0", "commercial": "yes" },
"downloads_30d": 134174,
"vram_estimates": [
{ "quant": "Q4_K_M", "gb": 25.1 },
{ "quant": "Q5_K_M", "gb": 28.5 }
],
"updated_at": "2026-11-10T01:00:51Z"
}
Specifications
- Parameters
- 30.0 B
- Licence
- apache-2.0
- First seen on the Hub
- 2026-10-16
- Training datasets
- undisclosed
- Added to our catalog
- 2026-11-10
Compare with any text-generation model