has anyone managed to get something coherent out of this model ? The few times I tried it seemed completely confused as if it was a much smaller model
ppsendo
2
Unfortunately, it works poorly; I tested it in claude-code, cowork, and my own application. It can’t handle it at all. I don’t know if it’s an issue with my configuration, but performance is weak. Other models, e.g., GLM, work very well.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Inkling-Small-NVFP4 on a Dual DGX Spark Cluster | 15 | 1714 | August 6, 2026 | |
| Inkling 952b and Inkling Small 276b | 1 | 1112 | July 17, 2026 | |
| Inkling Small | 18 | 1986 | August 22, 2026 | |
| LOT OF PROBLEMS CONNECTING / API INFERENCE | 0 | 113 | May 14, 2026 | |
| All new model are timeout | 3 | 560 | May 5, 2026 | |
| Remove Deepseek models and add GLM models again | 0 | 228 | September 2, 2026 | |
| Step-3.7-AWQ: 2xSpark: 48TG at C1, 108Toks at C8 | 0 | 425 | June 15, 2026 | |
| Keep GLM 5.3! | 7 | 721 | October 29, 2026 | |
| Ornith-1.0-397B Released – Has Anyone Tested It Yet or Found the Best Inference Settings? | 22 | 8462 | July 3, 2026 | |
| Ant Ling-3.0-Flash 124B-A5B... new fast model for one Spark! | 69 | 9408 | August 18, 2026 |