Just saw this on Intel’s huggingface repo. Haven’t had the time to try this yet. Their 122b autoround quant was pretty good so I have high hopes for this. Seems to be a good fit for 2x spark cluster.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| internlm/Intern-S2-Preview-397B about Claude Opus-4.8, GPT-5.5 performance | 1 | 452 | July 18, 2026 | |
| Intel GLM-5.3-Flash-W4A16-AutoRound | 7 | 964 | September 4, 2026 | |
| Qwen 3.5 SLM on DGX GB10 | 12 | 710 | March 3, 2026 | |
| OrcaRouter's Qwen3.8-Flash-Next-Uncensored on a Single DGX-Spark | 0 | 880 | September 1, 2026 | |
| Single DGX-Spark - Qwen 3.8-Flash-Next at ~43tok/sec in Coding | 63 | 10266 | September 14, 2026 | |
| Qwen/Qwen3.5-122B-A10B - Alibaba/Qwen thought about us... :-D | 340 | 19230 | March 24, 2026 | |
| Qwen3.5-122B-A10B NVFP4 Quantized for DGX Spark — 234GB → 75GB, Runs on 128GB | 44 | 13326 | April 9, 2026 | |
| Making your own Intel AutoRound quants on GB10 | 5 | 630 | March 13, 2026 | |
| Best Daily Single Spark Driver - King Model - Still Qwen 3.5 122b? | 15 | 2002 | July 27, 2026 | |
| Qwen3.5-397B-A17B run in dual spark! but I have a concern | 236 | 11019 | June 6, 2026 |