Eval history
Every EVAL_LOG=1 run, recorded. Two metrics, kept separate: retrieval hit@k (did the gold chunk reach the generator) vs answer accuracy (was the final answer right). A green retrieval cell with a ✗ means good retrieval but a wrong answer; a red cell with a ✓ usually means the model answered from parametric memory, not the document.
View as timeline → — the runs in chronological order with the story and deltas.
embedder:ollama:nomic-embed-textollama:qwen3-embedding:0.6bopenai:Qwen/Qwen3-Embedding-8Bopenai:jina-embeddings-v3 ✕
label:ab:chunkab:embedderab:plannerab:retrievalchunk:ctx-prefixdeploy:finaldeploy:pg-chunks_v2deploy:pg-genfixdeploy:pg-judgeddeploy:q8-sfdeploy:q8-sf-xdocfixrerank:jina-rawtextrerank:jina-v2rrf:vw1.0rrf:vw1.2tbl:f3-nomictbl:family3tbl:family3btbl:inmem
clear filtersRuns (19/25)
| When | Config | Label | git | Retrieval | Answer | n |
|---|---|---|---|---|---|---|
| 2026-06-25 10:51 | jina-v3 + deepseek-chat · k=20 | tbl:family3b | 16da4bb | 80% | 89% | 45 |
| 2026-06-25 09:57 | jina-v3 + deepseek-chat · k=20 | tbl:family3 | 16da4bb | 65% | 87% | 45 |
| 2026-06-23 10:06 | jina-v3 + deepseek-chat · k=20 | tbl:inmem | 16da4bb | 75% | 98% | 45 |
| 2026-06-23 09:02 | jina-v3 + deepseek-chat · k=20 | deploy:final | 3a94a4e | 83% | 98% | 45 |
| 2026-06-23 06:04 | jina-v3 + deepseek-chat · k=20 | deploy:pg-judged | 9169c2b | 80% | 98% | 45 |
| 2026-06-23 05:06 | jina-v3 + deepseek-chat · k=20 | deploy:pg-genfix | cd29729 | 80% | 98% | 45 |
| 2026-06-22 06:08 | jina-v3 + deepseek-chat · k=20 | deploy:pg-chunks_v2 | aa47cbb | 80% | 91% | 45 |
| 2026-06-22 05:20 | jina-v3 + deepseek-chat · k=20 | rerank:jina-rawtext | 805cec6 | 85% | 87% | 45 |
| 2026-06-22 04:44 | jina-v3 + deepseek-chat · k=20 | rerank:jina-v2 | 805cec6 | 73% | 80% | 45 |
| 2026-06-22 03:54 | jina-v3 + deepseek-chat · k=20 | chunk:ctx-prefix | 805cec6 | 70% | 80% | 45 |
| 2026-06-22 02:29 | jina-v3 + deepseek-chat · k=20 | rrf:vw1.2 | 94e4852 | 73% | 82% | 45 |
| 2026-06-22 02:27 | jina-v3 + deepseek-chat · k=20 | rrf:vw1.0 | 94e4852 | 70% | 82% | 45 |
| 2026-06-21 11:21 | jina-v3 + deepseek-chat · k=20 | ab:retrieval | 6025484 | 70% | 82% | 45 |
| 2026-06-21 11:18 | jina-v3 + deepseek-chat · k=20 | ab:retrieval | 6025484 | 63% | 78% | 45 |
| 2026-06-21 11:16 | jina-v3 + deepseek-chat · k=20 | ab:retrieval | 6025484 | 63% | 78% | 45 |
| 2026-06-21 09:58 | jina-v3 + deepseek-chat · k=20 | ab:planner | 1a133c4 | 63% | 80% | 46 |
| 2026-06-21 09:56 | jina-v3 + deepseek-chat · k=20 | ab:planner | 1a133c4 | 63% | 78% | 46 |
| 2026-06-19 08:23 | jina-v3 + deepseek-chat · k=20 | ab:chunk | 0562acf | 55% | 71% | 45 |
| 2026-06-18 16:21 | jina-v3 + deepseek-chat · k=20 | ab:embedder | c2bd01c | 63% | 78% | 45 |
Per-case grid
Cell colour = retrieval (■ hit, ■ miss, ■ absent). Glyph = answer (✓/✗).
| config | q001 | q002 | q003 | q004 | q005 | q006 | q007 | q008 | q009 | q010 | q011 | q012 | q013 | q014 | q015 | q016 | q017 | q018 | q019 | q020 | q021 | q022 | q023 | q024 | q025 | q026 | q027 | q028 | q029 | q030 | q031 | q032 | q033 | q034 | q035 | q036 | q037 | q038 | q039 | q040 | q041 | q042 | q043 | q044 | q045 | q046 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| jina-v3 + deepseek-chat · k=20 16da4bb | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✓ | ✓ | ✗ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 16da4bb | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✓ | ✗ | ✗ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 16da4bb | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 3a94a4e | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 9169c2b | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 cd29729 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 aa47cbb | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 805cec6 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 805cec6 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✗ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 805cec6 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 94e4852 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 94e4852 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 6025484 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 6025484 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✗ | ✓ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 6025484 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✗ | ✓ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 1a133c4 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ |
| jina-v3 + deepseek-chat · k=20 1a133c4 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✗ | ✓ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ |
| jina-v3 + deepseek-chat · k=20 0562acf | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | |
| jina-v3 + deepseek-chat · k=20 c2bd01c | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✗ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✗ | ✓ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ |