context-compressor
Public
bench / results / 2026-08-24-meta-llama-3.2-3b.md
bench / results / 2026-08-24-meta-llama-3.2-3b.md
Model: meta/llama-3.2-3b
Settings: {"limitTokens":500,"floorTokens":400,"keepRecentTokens":300,"chunkTokens":800,"summarizerMaxTokens":400,"consolidationBudgetTokens":600,"consolidationKeepNewest":2,"mergeMaxTokens":400,"maxExcerptChars":32000,"quizMaxTokens":120}
| Fixture | Recall (compacted) | Recall (consolidated) | Reduction | Chunk latency mean/p50/max (ms) | Structure | Cache |
|---|---|---|---|---|---|---|
| coding-agent | 4/5 | — | 33% | 1385.7/1408/1542 | ✅ | ✅ |
| research | 3/3 | — | 38% | 1085/1085/1085 | ✅ | ✅ |
| contradictory-updates | 3/3 | — | 23% | 666/666/666 | ✅ | ✅ |
| tool-spam | 3/3 | 3/3 | 56% | 871.6/590/1735 | ✅ | ✅ |
Recall = seeded facts answered correctly from the compacted view alone. Results are comparable only within one model; see bench/README.md.
Model: meta/llama-3.2-3b
Settings: {"limitTokens":500,"floorTokens":400,"keepRecentTokens":300,"chunkTokens":800,"summarizerMaxTokens":400,"consolidationBudgetTokens":600,"consolidationKeepNewest":2,"mergeMaxTokens":400,"maxExcerptChars":32000,"quizMaxTokens":120}
| Fixture | Recall (compacted) | Recall (consolidated) | Reduction | Chunk latency mean/p50/max (ms) | Structure | Cache |
|---|---|---|---|---|---|---|
| coding-agent | 4/5 | — | 33% | 1385.7/1408/1542 | ✅ | ✅ |
| research | 3/3 | — | 38% | 1085/1085/1085 | ✅ | ✅ |
| contradictory-updates | 3/3 | — | 23% | 666/666/666 | ✅ | ✅ |
| tool-spam | 3/3 | 3/3 | 56% | 871.6/590/1735 | ✅ | ✅ |
Recall = seeded facts answered correctly from the compacted view alone. Results are comparable only within one model; see bench/README.md.