dev-agent created branch executor-qwen3.8-27b in dev/nisse 2026-09-29 14:18:52 +00:00
dev-agent created pull request dev/nisse#56 2026-09-29 14:18:50 +00:00
Use qwen3.8-27b as the example executor
dev-agent created pull request dev/nisse#55 2026-09-28 03:15:04 +00:00
Benchmark the finalists on all eight tasks, and archive older results
dev-agent created branch bench-results-harder-tasks in dev/nisse 2026-09-28 03:14:42 +00:00
dev-agent pushed to bench-results-harder-tasks at dev/nisse 2026-09-28 03:14:42 +00:00
4b6c48cc96 Benchmark the finalists on all eight tasks, and archive older results
dev-agent pushed to spec-shared-inference-host at dev/nisse 2026-09-28 01:32:23 +00:00
2ab86bf0da Spec: OpenSCAD work needs vision, in nisse-dev or its own agent
dev-agent pushed to spec-shared-inference-host at dev/nisse 2026-09-28 01:30:37 +00:00
75d96b4197 Spec: restore the blank line before the section rule
dev-agent pushed to spec-shared-inference-host at dev/nisse 2026-09-28 01:30:21 +00:00
9239a56bd1 Spec: park vision as a need of future agents, not current ones
dev-agent pushed to spec-shared-inference-host at dev/nisse 2026-09-28 01:29:32 +00:00
50a2a75342 Spec: prefer one local model; add vision as an open decision
dev-agent pushed to spec-shared-inference-host at dev/nisse 2026-09-28 01:25:40 +00:00
e5c18b3bf4 Spec: record sharing the local inference host as an open decision
dev-agent created pull request dev/nisse#54 2026-09-28 01:25:39 +00:00
Spec: record sharing the local inference host as an open decision
dev-agent created branch spec-shared-inference-host in dev/nisse 2026-09-28 01:25:39 +00:00
dev-agent pushed to main at dev/nisse 2026-09-28 00:31:18 +00:00
8467b8747f Merge pull request 'Add four harder benchmark tasks' (#53) from harder-bench-tasks into main
277bff6fcd trace-bug: say how much precision the rates need
25fd57351d Add four harder benchmark tasks
Compare 3 commits »
dev-agent deleted branch harder-bench-tasks from dev/nisse 2026-09-28 00:31:18 +00:00
dev-agent merged pull request dev/nisse#53 2026-09-28 00:31:17 +00:00
Add four harder benchmark tasks
dev-agent created pull request dev/nisse#53 2026-09-28 00:08:53 +00:00
Add four harder benchmark tasks
dev-agent created branch harder-bench-tasks in dev/nisse 2026-09-28 00:08:53 +00:00
dev-agent pushed to harder-bench-tasks at dev/nisse 2026-09-28 00:08:53 +00:00
277bff6fcd trace-bug: say how much precision the rates need
25fd57351d Add four harder benchmark tasks
Compare 2 commits »
dev-agent pushed to report-keeps-tuning-apart at dev/nisse 2026-09-27 22:25:40 +00:00
f881a18ca4 Keep runs with different tuning apart in the report, and raise max_turns to 40
a77c84c128 Merge pull request 'Render one assistant message per turn, and make the request timeout a setting' (#51) from fix-transcript-and-timeout into main
d49435d4e5 Render one assistant message per turn, and make the request timeout a setting
Compare 3 commits »
dev-agent deleted branch fix-transcript-and-timeout from dev/nisse 2026-09-27 22:24:51 +00:00