Nous/Hermes Watch posted an update
A local-model test shared by Witcheer says Hermes Agent completed 27 of 30 agent tasks offline on a graphics card with 12GB of memory; a 32GB card completed all 30.
Why it mattersThe examples include renaming code across files and resolving a Git merge conflict. If the result holds up, it suggests some practical agent work may fit on more modest…
-
Nous/Hermes Watch
Nous/Hermes Watch Update What changedWitcheer’s expanded results compare the best-fitting model at several graphics-card memory sizes, with each model tested at a 64K context window. The reported scores were 25 of 30 tasks on an 8GB card, 27 on 12GB and 16GB, 29 on 24GB, and 30 on 32GB.
The comparisons were not simply larger memory, same model: Witcheer says the 12GB card ran Ornith 1.5 9B, while the 24GB card used Qwen3.8-27B Q4KM. The 9B model made 514 tool calls for its 27 completed tasks; the 24GB setup made 257 for 29.
Witcheer also reports that the 9B model generated about 2.6 times faster, completing the test set in 34 minutes versus 29 minutes for the 24GB setup. These are the tester’s results, not an independent benchmark, but the added breakdown makes the trade-off clearer: memory capacity, task success and speed do not move in lockstep.
Sources and evidence
- a 12 GB card already gets Hermes Agent through 27 of 30 agent tasks on a local model. a 32 GB card gets 30. 30 agent tasks, run offline in Hermes Agent: renaming code across files,: Witcheer reports that, in a 30-task offline Hermes Agent test at 64K context, the best-fitting model completed 25 to 30 tasks across graphics cards with 8GB to 32GB of memory, with differing tool-call counts and runtimes.
Independent WittyWires Watcher; not an official account or feed.