bench / results / 2026-08-24-openai-gpt-oss-20b.json
bench / results / 2026-08-24-openai-gpt-oss-20b.json
{
"date": "2026-08-24",
"modelId": "openai/gpt-oss-20b",
"variant": "full",
"settings": {
"limitTokens": 500,
"floorTokens": 400,
"keepRecentTokens": 300,
"chunkTokens": 800,
"summarizerMaxTokens": 400,
"consolidationBudgetTokens": 600,
"consolidationKeepNewest": 2,
"mergeMaxTokens": 400,
"maxExcerptChars": 32000,
"quizMaxTokens": 120
},
"fixtures": [
{
"fixture": "coding-agent",
"asked": 5,
"correctAfterCompaction": 5,
"reductionPercent": 33,
"latency": {
"mean": 1739.7,
"p50": 1617,
"max": 2045
},
"structural": {
"singleLeadingSystem": true,
"noOrphanToolResults": true,
"hasUserMessage": true,
"ok": true
},
"cacheDeterministic": true
},
{
"fixture": "research",
"asked": 3,
"correctAfterCompaction": 2,
"reductionPercent": 34,
"latency": {
"mean": 1453,
"p50": 1453,
"max": 1453
},
"structural": {
"singleLeadingSystem": true,
"noOrphanToolResults": true,
"hasUserMessage": true,
"ok": true
},
"cacheDeterministic": true
},
{
"fixture": "contradictory-updates",
"asked": 3,
"correctAfterCompaction": 3,
"reductionPercent": 10,
"latency": {
"mean": 1251,
"p50": 1251,
"max": 1251
},
"structural": {
"singleLeadingSystem": true,
"noOrphanToolResults": true,
"hasUserMessage": true,
"ok": true
},
"cacheDeterministic": true
},
{
"fixture": "tool-spam",
"asked": 3,
"correctAfterCompaction": 1,
"correctAfterConsolidation": 1,
"reductionPercent": 61,
"latency": {
"mean": 1131.4,
"p50": 1132,
"max": 1639
},
"structural": {
"singleLeadingSystem": true,
"noOrphanToolResults": true,
"hasUserMessage": true,
"ok": true
},
"cacheDeterministic": true
}
]
}{
"date": "2026-08-24",
"modelId": "openai/gpt-oss-20b",
"variant": "full",
"settings": {
"limitTokens": 500,
"floorTokens": 400,
"keepRecentTokens": 300,
"chunkTokens": 800,
"summarizerMaxTokens": 400,
"consolidationBudgetTokens": 600,
"consolidationKeepNewest": 2,
"mergeMaxTokens": 400,
"maxExcerptChars": 32000,
"quizMaxTokens": 120
},
"fixtures": [
{
"fixture": "coding-agent",
"asked": 5,
"correctAfterCompaction": 5,
"reductionPercent": 33,
"latency": {
"mean": 1739.7,
"p50": 1617,
"max": 2045
},
"structural": {
"singleLeadingSystem": true,
"noOrphanToolResults": true,
"hasUserMessage": true,
"ok": true
},
"cacheDeterministic": true
},
{
"fixture": "research",
"asked": 3,
"correctAfterCompaction": 2,
"reductionPercent": 34,
"latency": {
"mean": 1453,
"p50": 1453,
"max": 1453
},
"structural": {
"singleLeadingSystem": true,
"noOrphanToolResults": true,
"hasUserMessage": true,
"ok": true
},
"cacheDeterministic": true
},
{
"fixture": "contradictory-updates",
"asked": 3,
"correctAfterCompaction": 3,
"reductionPercent": 10,
"latency": {
"mean": 1251,
"p50": 1251,
"max": 1251
},
"structural": {
"singleLeadingSystem": true,
"noOrphanToolResults": true,
"hasUserMessage": true,
"ok": true
},
"cacheDeterministic": true
},
{
"fixture": "tool-spam",
"asked": 3,
"correctAfterCompaction": 1,
"correctAfterConsolidation": 1,
"reductionPercent": 61,
"latency": {
"mean": 1131.4,
"p50": 1132,
"max": 1639
},
"structural": {
"singleLeadingSystem": true,
"noOrphanToolResults": true,
"hasUserMessage": true,
"ok": true
},
"cacheDeterministic": true
}
]
}