@dotobokuri/fleet-console 1.57.1 → 1.57.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/dist/cli.mjs
CHANGED
|
@@ -33483,7 +33483,7 @@ var benchmarks_default = {
|
|
|
33483
33483
|
cursorbench: {
|
|
33484
33484
|
name: "CursorBench",
|
|
33485
33485
|
benchVersion: "3.2",
|
|
33486
|
-
observedAt: "2026-
|
|
33486
|
+
observedAt: "2026-08-13T00:00:00Z",
|
|
33487
33487
|
url: "https://cursor.com/cursorbench",
|
|
33488
33488
|
method: "Third-party agentic-coding benchmark: ambiguous multi-file tasks from real Cursor sessions, run in Cursor's own harness. Scores are evidence about the vendor model and transfer across gateway providers serving it; token and step figures are harness-relative efficiency signals, not billing facts.",
|
|
33489
33489
|
routingTieBandPoints: 2
|
|
@@ -33601,6 +33601,31 @@ var benchmarks_default = {
|
|
|
33601
33601
|
},
|
|
33602
33602
|
caveat: "Bench footnote: an earlier snapshot of Cursor's codebase was unintentionally present in this model's training data, inflating scores by an unknown amount; the advantage does not transfer to other repositories."
|
|
33603
33603
|
},
|
|
33604
|
+
"grok-4.6": {
|
|
33605
|
+
source: "cursorbench",
|
|
33606
|
+
rungs: {
|
|
33607
|
+
low: {
|
|
33608
|
+
score: 61,
|
|
33609
|
+
tokensPerTask: 10658,
|
|
33610
|
+
stepsPerTask: 23
|
|
33611
|
+
},
|
|
33612
|
+
medium: {
|
|
33613
|
+
score: 67.1,
|
|
33614
|
+
tokensPerTask: 17942,
|
|
33615
|
+
stepsPerTask: 29
|
|
33616
|
+
},
|
|
33617
|
+
high: {
|
|
33618
|
+
score: 69.9,
|
|
33619
|
+
tokensPerTask: 32449,
|
|
33620
|
+
stepsPerTask: 39
|
|
33621
|
+
},
|
|
33622
|
+
xhigh: {
|
|
33623
|
+
score: 70.8,
|
|
33624
|
+
tokensPerTask: 41136,
|
|
33625
|
+
stepsPerTask: 46
|
|
33626
|
+
}
|
|
33627
|
+
}
|
|
33628
|
+
},
|
|
33604
33629
|
"kimi-k3": {
|
|
33605
33630
|
source: "cursorbench",
|
|
33606
33631
|
rungs: {
|
|
@@ -33649,7 +33674,7 @@ var benchmarks_default = {
|
|
|
33649
33674
|
};
|
|
33650
33675
|
var models_default = {
|
|
33651
33676
|
version: 1,
|
|
33652
|
-
updatedAt: "2026-08-
|
|
33677
|
+
updatedAt: "2026-08-13T00:00:00Z",
|
|
33653
33678
|
providers: {
|
|
33654
33679
|
codex: {
|
|
33655
33680
|
name: "Codex",
|
|
@@ -33791,7 +33816,7 @@ var models_default = {
|
|
|
33791
33816
|
cursor: {
|
|
33792
33817
|
name: "Cursor",
|
|
33793
33818
|
defaultModel: "auto",
|
|
33794
|
-
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
33819
|
+
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); cursor-agent 2026.08.11-e8db854 observed 2026-08-13 for grok-4.6; quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
33795
33820
|
models: [
|
|
33796
33821
|
{
|
|
33797
33822
|
modelId: "auto",
|
|
@@ -33850,6 +33875,41 @@ var models_default = {
|
|
|
33850
33875
|
],
|
|
33851
33876
|
upstreamModelIdTemplate: "cursor-grok-4.5-{effort}-fast"
|
|
33852
33877
|
}
|
|
33878
|
+
},
|
|
33879
|
+
{
|
|
33880
|
+
modelId: "grok-4.6",
|
|
33881
|
+
name: "Grok-4.6",
|
|
33882
|
+
capabilityClass: "flagship",
|
|
33883
|
+
quotaScope: "auto",
|
|
33884
|
+
contextWindow: 256e3,
|
|
33885
|
+
effort: {
|
|
33886
|
+
supported: true,
|
|
33887
|
+
levels: [
|
|
33888
|
+
"low",
|
|
33889
|
+
"medium",
|
|
33890
|
+
"high",
|
|
33891
|
+
"xhigh"
|
|
33892
|
+
],
|
|
33893
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}"
|
|
33894
|
+
},
|
|
33895
|
+
benchmarkKey: "grok-4.6"
|
|
33896
|
+
},
|
|
33897
|
+
{
|
|
33898
|
+
modelId: "grok-4.6-fast",
|
|
33899
|
+
name: "Grok-4.6-Fast",
|
|
33900
|
+
capabilityClass: "light",
|
|
33901
|
+
quotaScope: "auto",
|
|
33902
|
+
contextWindow: 256e3,
|
|
33903
|
+
effort: {
|
|
33904
|
+
supported: true,
|
|
33905
|
+
levels: [
|
|
33906
|
+
"low",
|
|
33907
|
+
"medium",
|
|
33908
|
+
"high",
|
|
33909
|
+
"xhigh"
|
|
33910
|
+
],
|
|
33911
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}-fast"
|
|
33912
|
+
}
|
|
33853
33913
|
}
|
|
33854
33914
|
]
|
|
33855
33915
|
},
|
|
@@ -17820,7 +17820,7 @@ var benchmarks_default = {
|
|
|
17820
17820
|
cursorbench: {
|
|
17821
17821
|
name: "CursorBench",
|
|
17822
17822
|
benchVersion: "3.2",
|
|
17823
|
-
observedAt: "2026-
|
|
17823
|
+
observedAt: "2026-08-13T00:00:00Z",
|
|
17824
17824
|
url: "https://cursor.com/cursorbench",
|
|
17825
17825
|
method: "Third-party agentic-coding benchmark: ambiguous multi-file tasks from real Cursor sessions, run in Cursor's own harness. Scores are evidence about the vendor model and transfer across gateway providers serving it; token and step figures are harness-relative efficiency signals, not billing facts.",
|
|
17826
17826
|
routingTieBandPoints: 2
|
|
@@ -17938,6 +17938,31 @@ var benchmarks_default = {
|
|
|
17938
17938
|
},
|
|
17939
17939
|
caveat: "Bench footnote: an earlier snapshot of Cursor's codebase was unintentionally present in this model's training data, inflating scores by an unknown amount; the advantage does not transfer to other repositories."
|
|
17940
17940
|
},
|
|
17941
|
+
"grok-4.6": {
|
|
17942
|
+
source: "cursorbench",
|
|
17943
|
+
rungs: {
|
|
17944
|
+
low: {
|
|
17945
|
+
score: 61,
|
|
17946
|
+
tokensPerTask: 10658,
|
|
17947
|
+
stepsPerTask: 23
|
|
17948
|
+
},
|
|
17949
|
+
medium: {
|
|
17950
|
+
score: 67.1,
|
|
17951
|
+
tokensPerTask: 17942,
|
|
17952
|
+
stepsPerTask: 29
|
|
17953
|
+
},
|
|
17954
|
+
high: {
|
|
17955
|
+
score: 69.9,
|
|
17956
|
+
tokensPerTask: 32449,
|
|
17957
|
+
stepsPerTask: 39
|
|
17958
|
+
},
|
|
17959
|
+
xhigh: {
|
|
17960
|
+
score: 70.8,
|
|
17961
|
+
tokensPerTask: 41136,
|
|
17962
|
+
stepsPerTask: 46
|
|
17963
|
+
}
|
|
17964
|
+
}
|
|
17965
|
+
},
|
|
17941
17966
|
"kimi-k3": {
|
|
17942
17967
|
source: "cursorbench",
|
|
17943
17968
|
rungs: {
|
|
@@ -17986,7 +18011,7 @@ var benchmarks_default = {
|
|
|
17986
18011
|
};
|
|
17987
18012
|
var models_default = {
|
|
17988
18013
|
version: 1,
|
|
17989
|
-
updatedAt: "2026-08-
|
|
18014
|
+
updatedAt: "2026-08-13T00:00:00Z",
|
|
17990
18015
|
providers: {
|
|
17991
18016
|
codex: {
|
|
17992
18017
|
name: "Codex",
|
|
@@ -18128,7 +18153,7 @@ var models_default = {
|
|
|
18128
18153
|
cursor: {
|
|
18129
18154
|
name: "Cursor",
|
|
18130
18155
|
defaultModel: "auto",
|
|
18131
|
-
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
18156
|
+
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); cursor-agent 2026.08.11-e8db854 observed 2026-08-13 for grok-4.6; quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
18132
18157
|
models: [
|
|
18133
18158
|
{
|
|
18134
18159
|
modelId: "auto",
|
|
@@ -18187,6 +18212,41 @@ var models_default = {
|
|
|
18187
18212
|
],
|
|
18188
18213
|
upstreamModelIdTemplate: "cursor-grok-4.5-{effort}-fast"
|
|
18189
18214
|
}
|
|
18215
|
+
},
|
|
18216
|
+
{
|
|
18217
|
+
modelId: "grok-4.6",
|
|
18218
|
+
name: "Grok-4.6",
|
|
18219
|
+
capabilityClass: "flagship",
|
|
18220
|
+
quotaScope: "auto",
|
|
18221
|
+
contextWindow: 256e3,
|
|
18222
|
+
effort: {
|
|
18223
|
+
supported: true,
|
|
18224
|
+
levels: [
|
|
18225
|
+
"low",
|
|
18226
|
+
"medium",
|
|
18227
|
+
"high",
|
|
18228
|
+
"xhigh"
|
|
18229
|
+
],
|
|
18230
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}"
|
|
18231
|
+
},
|
|
18232
|
+
benchmarkKey: "grok-4.6"
|
|
18233
|
+
},
|
|
18234
|
+
{
|
|
18235
|
+
modelId: "grok-4.6-fast",
|
|
18236
|
+
name: "Grok-4.6-Fast",
|
|
18237
|
+
capabilityClass: "light",
|
|
18238
|
+
quotaScope: "auto",
|
|
18239
|
+
contextWindow: 256e3,
|
|
18240
|
+
effort: {
|
|
18241
|
+
supported: true,
|
|
18242
|
+
levels: [
|
|
18243
|
+
"low",
|
|
18244
|
+
"medium",
|
|
18245
|
+
"high",
|
|
18246
|
+
"xhigh"
|
|
18247
|
+
],
|
|
18248
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}-fast"
|
|
18249
|
+
}
|
|
18190
18250
|
}
|
|
18191
18251
|
]
|
|
18192
18252
|
},
|
|
@@ -17494,7 +17494,7 @@ var benchmarks_default = {
|
|
|
17494
17494
|
cursorbench: {
|
|
17495
17495
|
name: "CursorBench",
|
|
17496
17496
|
benchVersion: "3.2",
|
|
17497
|
-
observedAt: "2026-
|
|
17497
|
+
observedAt: "2026-08-13T00:00:00Z",
|
|
17498
17498
|
url: "https://cursor.com/cursorbench",
|
|
17499
17499
|
method: "Third-party agentic-coding benchmark: ambiguous multi-file tasks from real Cursor sessions, run in Cursor's own harness. Scores are evidence about the vendor model and transfer across gateway providers serving it; token and step figures are harness-relative efficiency signals, not billing facts.",
|
|
17500
17500
|
routingTieBandPoints: 2
|
|
@@ -17612,6 +17612,31 @@ var benchmarks_default = {
|
|
|
17612
17612
|
},
|
|
17613
17613
|
caveat: "Bench footnote: an earlier snapshot of Cursor's codebase was unintentionally present in this model's training data, inflating scores by an unknown amount; the advantage does not transfer to other repositories."
|
|
17614
17614
|
},
|
|
17615
|
+
"grok-4.6": {
|
|
17616
|
+
source: "cursorbench",
|
|
17617
|
+
rungs: {
|
|
17618
|
+
low: {
|
|
17619
|
+
score: 61,
|
|
17620
|
+
tokensPerTask: 10658,
|
|
17621
|
+
stepsPerTask: 23
|
|
17622
|
+
},
|
|
17623
|
+
medium: {
|
|
17624
|
+
score: 67.1,
|
|
17625
|
+
tokensPerTask: 17942,
|
|
17626
|
+
stepsPerTask: 29
|
|
17627
|
+
},
|
|
17628
|
+
high: {
|
|
17629
|
+
score: 69.9,
|
|
17630
|
+
tokensPerTask: 32449,
|
|
17631
|
+
stepsPerTask: 39
|
|
17632
|
+
},
|
|
17633
|
+
xhigh: {
|
|
17634
|
+
score: 70.8,
|
|
17635
|
+
tokensPerTask: 41136,
|
|
17636
|
+
stepsPerTask: 46
|
|
17637
|
+
}
|
|
17638
|
+
}
|
|
17639
|
+
},
|
|
17615
17640
|
"kimi-k3": {
|
|
17616
17641
|
source: "cursorbench",
|
|
17617
17642
|
rungs: {
|
|
@@ -17660,7 +17685,7 @@ var benchmarks_default = {
|
|
|
17660
17685
|
};
|
|
17661
17686
|
var models_default = {
|
|
17662
17687
|
version: 1,
|
|
17663
|
-
updatedAt: "2026-08-
|
|
17688
|
+
updatedAt: "2026-08-13T00:00:00Z",
|
|
17664
17689
|
providers: {
|
|
17665
17690
|
codex: {
|
|
17666
17691
|
name: "Codex",
|
|
@@ -17802,7 +17827,7 @@ var models_default = {
|
|
|
17802
17827
|
cursor: {
|
|
17803
17828
|
name: "Cursor",
|
|
17804
17829
|
defaultModel: "auto",
|
|
17805
|
-
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
17830
|
+
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); cursor-agent 2026.08.11-e8db854 observed 2026-08-13 for grok-4.6; quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
17806
17831
|
models: [
|
|
17807
17832
|
{
|
|
17808
17833
|
modelId: "auto",
|
|
@@ -17861,6 +17886,41 @@ var models_default = {
|
|
|
17861
17886
|
],
|
|
17862
17887
|
upstreamModelIdTemplate: "cursor-grok-4.5-{effort}-fast"
|
|
17863
17888
|
}
|
|
17889
|
+
},
|
|
17890
|
+
{
|
|
17891
|
+
modelId: "grok-4.6",
|
|
17892
|
+
name: "Grok-4.6",
|
|
17893
|
+
capabilityClass: "flagship",
|
|
17894
|
+
quotaScope: "auto",
|
|
17895
|
+
contextWindow: 256e3,
|
|
17896
|
+
effort: {
|
|
17897
|
+
supported: true,
|
|
17898
|
+
levels: [
|
|
17899
|
+
"low",
|
|
17900
|
+
"medium",
|
|
17901
|
+
"high",
|
|
17902
|
+
"xhigh"
|
|
17903
|
+
],
|
|
17904
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}"
|
|
17905
|
+
},
|
|
17906
|
+
benchmarkKey: "grok-4.6"
|
|
17907
|
+
},
|
|
17908
|
+
{
|
|
17909
|
+
modelId: "grok-4.6-fast",
|
|
17910
|
+
name: "Grok-4.6-Fast",
|
|
17911
|
+
capabilityClass: "light",
|
|
17912
|
+
quotaScope: "auto",
|
|
17913
|
+
contextWindow: 256e3,
|
|
17914
|
+
effort: {
|
|
17915
|
+
supported: true,
|
|
17916
|
+
levels: [
|
|
17917
|
+
"low",
|
|
17918
|
+
"medium",
|
|
17919
|
+
"high",
|
|
17920
|
+
"xhigh"
|
|
17921
|
+
],
|
|
17922
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}-fast"
|
|
17923
|
+
}
|
|
17864
17924
|
}
|
|
17865
17925
|
]
|
|
17866
17926
|
},
|
|
@@ -20492,7 +20492,7 @@ var benchmarks_default = {
|
|
|
20492
20492
|
cursorbench: {
|
|
20493
20493
|
name: "CursorBench",
|
|
20494
20494
|
benchVersion: "3.2",
|
|
20495
|
-
observedAt: "2026-
|
|
20495
|
+
observedAt: "2026-08-13T00:00:00Z",
|
|
20496
20496
|
url: "https://cursor.com/cursorbench",
|
|
20497
20497
|
method: "Third-party agentic-coding benchmark: ambiguous multi-file tasks from real Cursor sessions, run in Cursor's own harness. Scores are evidence about the vendor model and transfer across gateway providers serving it; token and step figures are harness-relative efficiency signals, not billing facts.",
|
|
20498
20498
|
routingTieBandPoints: 2
|
|
@@ -20610,6 +20610,31 @@ var benchmarks_default = {
|
|
|
20610
20610
|
},
|
|
20611
20611
|
caveat: "Bench footnote: an earlier snapshot of Cursor's codebase was unintentionally present in this model's training data, inflating scores by an unknown amount; the advantage does not transfer to other repositories."
|
|
20612
20612
|
},
|
|
20613
|
+
"grok-4.6": {
|
|
20614
|
+
source: "cursorbench",
|
|
20615
|
+
rungs: {
|
|
20616
|
+
low: {
|
|
20617
|
+
score: 61,
|
|
20618
|
+
tokensPerTask: 10658,
|
|
20619
|
+
stepsPerTask: 23
|
|
20620
|
+
},
|
|
20621
|
+
medium: {
|
|
20622
|
+
score: 67.1,
|
|
20623
|
+
tokensPerTask: 17942,
|
|
20624
|
+
stepsPerTask: 29
|
|
20625
|
+
},
|
|
20626
|
+
high: {
|
|
20627
|
+
score: 69.9,
|
|
20628
|
+
tokensPerTask: 32449,
|
|
20629
|
+
stepsPerTask: 39
|
|
20630
|
+
},
|
|
20631
|
+
xhigh: {
|
|
20632
|
+
score: 70.8,
|
|
20633
|
+
tokensPerTask: 41136,
|
|
20634
|
+
stepsPerTask: 46
|
|
20635
|
+
}
|
|
20636
|
+
}
|
|
20637
|
+
},
|
|
20613
20638
|
"kimi-k3": {
|
|
20614
20639
|
source: "cursorbench",
|
|
20615
20640
|
rungs: {
|
|
@@ -20658,7 +20683,7 @@ var benchmarks_default = {
|
|
|
20658
20683
|
};
|
|
20659
20684
|
var models_default = {
|
|
20660
20685
|
version: 1,
|
|
20661
|
-
updatedAt: "2026-08-
|
|
20686
|
+
updatedAt: "2026-08-13T00:00:00Z",
|
|
20662
20687
|
providers: {
|
|
20663
20688
|
codex: {
|
|
20664
20689
|
name: "Codex",
|
|
@@ -20800,7 +20825,7 @@ var models_default = {
|
|
|
20800
20825
|
cursor: {
|
|
20801
20826
|
name: "Cursor",
|
|
20802
20827
|
defaultModel: "auto",
|
|
20803
|
-
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
20828
|
+
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); cursor-agent 2026.08.11-e8db854 observed 2026-08-13 for grok-4.6; quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
20804
20829
|
models: [
|
|
20805
20830
|
{
|
|
20806
20831
|
modelId: "auto",
|
|
@@ -20859,6 +20884,41 @@ var models_default = {
|
|
|
20859
20884
|
],
|
|
20860
20885
|
upstreamModelIdTemplate: "cursor-grok-4.5-{effort}-fast"
|
|
20861
20886
|
}
|
|
20887
|
+
},
|
|
20888
|
+
{
|
|
20889
|
+
modelId: "grok-4.6",
|
|
20890
|
+
name: "Grok-4.6",
|
|
20891
|
+
capabilityClass: "flagship",
|
|
20892
|
+
quotaScope: "auto",
|
|
20893
|
+
contextWindow: 256e3,
|
|
20894
|
+
effort: {
|
|
20895
|
+
supported: true,
|
|
20896
|
+
levels: [
|
|
20897
|
+
"low",
|
|
20898
|
+
"medium",
|
|
20899
|
+
"high",
|
|
20900
|
+
"xhigh"
|
|
20901
|
+
],
|
|
20902
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}"
|
|
20903
|
+
},
|
|
20904
|
+
benchmarkKey: "grok-4.6"
|
|
20905
|
+
},
|
|
20906
|
+
{
|
|
20907
|
+
modelId: "grok-4.6-fast",
|
|
20908
|
+
name: "Grok-4.6-Fast",
|
|
20909
|
+
capabilityClass: "light",
|
|
20910
|
+
quotaScope: "auto",
|
|
20911
|
+
contextWindow: 256e3,
|
|
20912
|
+
effort: {
|
|
20913
|
+
supported: true,
|
|
20914
|
+
levels: [
|
|
20915
|
+
"low",
|
|
20916
|
+
"medium",
|
|
20917
|
+
"high",
|
|
20918
|
+
"xhigh"
|
|
20919
|
+
],
|
|
20920
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}-fast"
|
|
20921
|
+
}
|
|
20862
20922
|
}
|
|
20863
20923
|
]
|
|
20864
20924
|
},
|
package/dist/fleet.mjs
CHANGED
|
@@ -21907,7 +21907,7 @@ var benchmarks_default = {
|
|
|
21907
21907
|
cursorbench: {
|
|
21908
21908
|
name: "CursorBench",
|
|
21909
21909
|
benchVersion: "3.2",
|
|
21910
|
-
observedAt: "2026-
|
|
21910
|
+
observedAt: "2026-08-13T00:00:00Z",
|
|
21911
21911
|
url: "https://cursor.com/cursorbench",
|
|
21912
21912
|
method: "Third-party agentic-coding benchmark: ambiguous multi-file tasks from real Cursor sessions, run in Cursor's own harness. Scores are evidence about the vendor model and transfer across gateway providers serving it; token and step figures are harness-relative efficiency signals, not billing facts.",
|
|
21913
21913
|
routingTieBandPoints: 2
|
|
@@ -22025,6 +22025,31 @@ var benchmarks_default = {
|
|
|
22025
22025
|
},
|
|
22026
22026
|
caveat: "Bench footnote: an earlier snapshot of Cursor's codebase was unintentionally present in this model's training data, inflating scores by an unknown amount; the advantage does not transfer to other repositories."
|
|
22027
22027
|
},
|
|
22028
|
+
"grok-4.6": {
|
|
22029
|
+
source: "cursorbench",
|
|
22030
|
+
rungs: {
|
|
22031
|
+
low: {
|
|
22032
|
+
score: 61,
|
|
22033
|
+
tokensPerTask: 10658,
|
|
22034
|
+
stepsPerTask: 23
|
|
22035
|
+
},
|
|
22036
|
+
medium: {
|
|
22037
|
+
score: 67.1,
|
|
22038
|
+
tokensPerTask: 17942,
|
|
22039
|
+
stepsPerTask: 29
|
|
22040
|
+
},
|
|
22041
|
+
high: {
|
|
22042
|
+
score: 69.9,
|
|
22043
|
+
tokensPerTask: 32449,
|
|
22044
|
+
stepsPerTask: 39
|
|
22045
|
+
},
|
|
22046
|
+
xhigh: {
|
|
22047
|
+
score: 70.8,
|
|
22048
|
+
tokensPerTask: 41136,
|
|
22049
|
+
stepsPerTask: 46
|
|
22050
|
+
}
|
|
22051
|
+
}
|
|
22052
|
+
},
|
|
22028
22053
|
"kimi-k3": {
|
|
22029
22054
|
source: "cursorbench",
|
|
22030
22055
|
rungs: {
|
|
@@ -22073,7 +22098,7 @@ var benchmarks_default = {
|
|
|
22073
22098
|
};
|
|
22074
22099
|
var models_default = {
|
|
22075
22100
|
version: 1,
|
|
22076
|
-
updatedAt: "2026-08-
|
|
22101
|
+
updatedAt: "2026-08-13T00:00:00Z",
|
|
22077
22102
|
providers: {
|
|
22078
22103
|
codex: {
|
|
22079
22104
|
name: "Codex",
|
|
@@ -22215,7 +22240,7 @@ var models_default = {
|
|
|
22215
22240
|
cursor: {
|
|
22216
22241
|
name: "Cursor",
|
|
22217
22242
|
defaultModel: "auto",
|
|
22218
|
-
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
22243
|
+
source: "Cursor GetUsableModels + Run conversationCheckpointUpdate.token_details.max_tokens (cli-2026.07.08-0c04a8a; auto, composer-2.5, and grok-4.5 families observed 2026-08-01); cursor-agent 2026.08.11-e8db854 observed 2026-08-13 for grok-4.6; quotaScope from DashboardService/GetCurrentPeriodUsage autoBucketModels (2026-08-05)",
|
|
22219
22244
|
models: [
|
|
22220
22245
|
{
|
|
22221
22246
|
modelId: "auto",
|
|
@@ -22274,6 +22299,41 @@ var models_default = {
|
|
|
22274
22299
|
],
|
|
22275
22300
|
upstreamModelIdTemplate: "cursor-grok-4.5-{effort}-fast"
|
|
22276
22301
|
}
|
|
22302
|
+
},
|
|
22303
|
+
{
|
|
22304
|
+
modelId: "grok-4.6",
|
|
22305
|
+
name: "Grok-4.6",
|
|
22306
|
+
capabilityClass: "flagship",
|
|
22307
|
+
quotaScope: "auto",
|
|
22308
|
+
contextWindow: 256e3,
|
|
22309
|
+
effort: {
|
|
22310
|
+
supported: true,
|
|
22311
|
+
levels: [
|
|
22312
|
+
"low",
|
|
22313
|
+
"medium",
|
|
22314
|
+
"high",
|
|
22315
|
+
"xhigh"
|
|
22316
|
+
],
|
|
22317
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}"
|
|
22318
|
+
},
|
|
22319
|
+
benchmarkKey: "grok-4.6"
|
|
22320
|
+
},
|
|
22321
|
+
{
|
|
22322
|
+
modelId: "grok-4.6-fast",
|
|
22323
|
+
name: "Grok-4.6-Fast",
|
|
22324
|
+
capabilityClass: "light",
|
|
22325
|
+
quotaScope: "auto",
|
|
22326
|
+
contextWindow: 256e3,
|
|
22327
|
+
effort: {
|
|
22328
|
+
supported: true,
|
|
22329
|
+
levels: [
|
|
22330
|
+
"low",
|
|
22331
|
+
"medium",
|
|
22332
|
+
"high",
|
|
22333
|
+
"xhigh"
|
|
22334
|
+
],
|
|
22335
|
+
upstreamModelIdTemplate: "cursor-grok-4.6-{effort}-fast"
|
|
22336
|
+
}
|
|
22277
22337
|
}
|
|
22278
22338
|
]
|
|
22279
22339
|
},
|
package/package.json
CHANGED