mirror of
https://github.com/dotnet/skills.git
synced 2026-09-20 09:49:54 +08:00
Use MAI Code 1.1 for evaluations (#1174)
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
This commit is contained in:
@@ -1093,9 +1093,9 @@ jobs:
|
||||
# expensive and deliberately under-advertised — use sparingly.
|
||||
$profileModels = @{
|
||||
'default' = @('claude-sonnet-5', 'gpt-5.6-luna')
|
||||
'mid' = @('claude-haiku-4.5', 'mai-code-1-flash-picker', 'gpt-5.3-codex')
|
||||
'mid' = @('claude-haiku-4.5', 'mai-code-1.1-flash', 'gpt-5.3-codex')
|
||||
'opus48' = @('claude-opus-4.8')
|
||||
'full' = @('claude-sonnet-5', 'gpt-5.6-luna', 'claude-haiku-4.5', 'mai-code-1-flash-picker', 'gpt-5.3-codex', 'claude-opus-4.8')
|
||||
'full' = @('claude-sonnet-5', 'gpt-5.6-luna', 'claude-haiku-4.5', 'mai-code-1.1-flash', 'gpt-5.3-codex', 'claude-opus-4.8')
|
||||
'newer' = @('gpt-5.6-sol', 'claude-opus-5', 'claude-sonnet-5')
|
||||
}
|
||||
# §10.5 judge routing. Primary = cross-family (never the same model, and
|
||||
@@ -1110,7 +1110,7 @@ jobs:
|
||||
'claude-opus-4.8' = @{ judge = 'gpt-5.6-terra'; judge2 = '' }
|
||||
'gpt-5.6-luna' = @{ judge = 'claude-opus-4.8'; judge2 = 'claude-haiku-4.5' }
|
||||
'claude-haiku-4.5' = @{ judge = 'gpt-5.6-terra'; judge2 = '' }
|
||||
'mai-code-1-flash-picker' = @{ judge = 'gpt-5.6-terra'; judge2 = '' }
|
||||
'mai-code-1.1-flash' = @{ judge = 'gpt-5.6-terra'; judge2 = '' }
|
||||
'gpt-5.3-codex' = @{ judge = 'claude-opus-4.8'; judge2 = 'claude-haiku-4.5' }
|
||||
'gpt-5.6-sol' = @{ judge = 'claude-opus-4.8'; judge2 = 'claude-haiku-4.5' }
|
||||
'claude-sonnet-5' = @{ judge = 'gpt-5.6-terra'; judge2 = '' }
|
||||
|
||||
@@ -74,14 +74,14 @@ class TokenFailoverTests(unittest.TestCase):
|
||||
("issue_comment", "/evaluate", "", "", ["claude-sonnet-5", "gpt-5.6-luna"]),
|
||||
("pull_request_review", "/evaluate --full", "", "", [
|
||||
"claude-sonnet-5", "gpt-5.6-luna", "claude-haiku-4.5",
|
||||
"mai-code-1-flash-picker", "gpt-5.3-codex", "claude-opus-4.8",
|
||||
"mai-code-1.1-flash", "gpt-5.3-codex", "claude-opus-4.8",
|
||||
]),
|
||||
("workflow_dispatch", "", "newer", "", [
|
||||
"gpt-5.6-sol", "claude-opus-5", "claude-sonnet-5",
|
||||
]),
|
||||
("schedule", "", "", "0 7 * * 1,3,5", ["claude-sonnet-5", "gpt-5.6-luna"]),
|
||||
("schedule", "", "", "0 7 * * 2,6", [
|
||||
"claude-haiku-4.5", "mai-code-1-flash-picker", "gpt-5.3-codex",
|
||||
"claude-haiku-4.5", "mai-code-1.1-flash", "gpt-5.3-codex",
|
||||
]),
|
||||
("schedule", "", "", "0 7 * * 0", [
|
||||
"gpt-5.6-sol", "claude-opus-5", "claude-sonnet-5",
|
||||
|
||||
Reference in New Issue
Block a user