Suite results
Suite Score Avg / 20 Tests n tok/s Run
▶ General capability (13-task real-workload suite) 1a 213.4 / 260 19.4 11/13 ● n=1.1 17.6 run 2026-09-06
| Test | Category | Score | | tok/s |
| A1 | Dev/Ops Scripting | 17/20 | | 25.6 |
| A2 | Dev/Ops Scripting | 18/20 | | 23.3 |
| A4 | Dev/Ops Scripting | 19/20 | | 20.1 |
| A5 | Dev/Ops Scripting | 20/20 | | 9.2 |
| A6 | Dev/Ops Scripting | 20/20 | | 19.1 |
| B1 | Document Processing | 20/20 | | 13.9 |
| B2 | Document Processing | 20/20 | | 14.6 |
| C1 | Content Production | 20/20 | | 17 |
| C2 | Content Production | 18/20 | | 16.4 |
| C3 | Content Production | 20/20 | | 13.4 |
| C4 | Content Production | 20/20 | | 21.1 |
Category mean Dev/Ops Scripting 18.8Document Processing 20Content Production 19.5
▶ Agentic tool-calling & protocol adherence 1b 160 / 160 20 8/8 n=1 25.6 run 2026-09-06
| Test | Category | Score | | tok/s |
| WA1 | Tool Calling & Protocol Adherence | 20/20 | | 24.7 |
| WA2 | Tool Calling & Protocol Adherence | 20/20 | | 25.4 |
| WA3 | Tool Calling & Protocol Adherence | 20/20 | | 25.3 |
| WA4 | Tool Calling & Protocol Adherence | 20/20 | | 23.6 |
| WA5 | Tool Calling & Protocol Adherence | 20/20 | | 25.9 |
| WB1 | Agent Robustness & State Management | 20/20 | | 21.1 |
| WB2 | Agent Robustness & State Management | 20/20 | | 22.3 |
| WB3 | Agent Robustness & State Management | 20/20 | | 19.9 |
Category mean Tool Calling & Protocol Adherence 20Agent Robustness & State Management 20
▶ Coding depth 1c 176.4 / 180 19.6 9/9 n=1 18.8 run 2026-10-02
| Test | Category | Score | | tok/s |
| K1 | Coding — Concurrency | 20/20 | | 23.5 |
| K2 | Coding — Algorithms | 20/20 | | 20.8 |
| K3 | Coding — Security Review | 20/20 | | 18.7 |
| K4 | Coding — Performance | 19/20 | | 17.6 |
| K5 | Coding — Refactoring Under Constraints | 20/20 | | 17.2 |
| K6 | Coding — Test Writing | 20/20 | | 17.3 |
| K7 | Coding — React Component (typed) | 17/20 | | 17.6 |
| K8 | Coding — TypeScript Type System | 20/20 | | 17.5 |
| K9 | Coding — React Debugging | 20/20 | | 17.1 |
Category mean Coding — Concurrency 20Coding — Algorithms 20Coding — Security Review 20Coding — Performance 19Coding — Refactoring Under Constraints 20Coding — Test Writing 20Coding — React Component (typed) 17Coding — TypeScript Type System 20Coding — React Debugging 20
▶ Doc/OCR vision 1d 160 / 160 20 8/8 n=1 15.6 run 2026-09-28
| Test | Category | Score | | tok/s |
| D1 | Doc/OCR — Transcription | 20/20 | | 21.1 |
| D2 | Doc/OCR — Document QA | 20/20 | | 22.7 |
| D3 | Doc/OCR — Field Extraction | 20/20 | | 19 |
| D4 | Doc/OCR — Document QA | 20/20 | | 13 |
| D5 | Doc/OCR — Transcription | 20/20 | | 9.5 |
| D6 | Doc/OCR — Field Extraction | 20/20 | | 11.3 |
| D7 | Doc/OCR — Degraded Input | 20/20 | | 12.2 |
| D8 | Doc/OCR — Degraded Input | 20/20 | | 12.9 |
Category mean Doc/OCR — Transcription 20Doc/OCR — Document QA 20Doc/OCR — Field Extraction 20Doc/OCR — Degraded Input 20
▶ Doc/OCR — real-degraded tier 1d2 45 / 80 15 3/4 ● n=1 20.8 run 2026-09-22
| Test | Category | Score | | tok/s |
| D9 | Doc/OCR — Real Degraded Transcription | 9/20 | | 23.3 |
| D10 | Doc/OCR — Handwriting Extraction | 16/20 | | 20 |
| D12 | Doc/OCR — Degraded Table QA | 20/20 | | 19 |
Category mean Doc/OCR — Real Degraded Transcription 9Doc/OCR — Handwriting Extraction 16Doc/OCR — Degraded Table QA 20
▶ Content-production depth 1e 115.2 / 120 19.2 6/6 n=1 15.7 run 2026-09-30
| Test | Category | Score | | tok/s |
| E1 | SOW & Requirements Decomposition | 20/20 | | 19.5 |
| E2 | SOW & Requirements Decomposition | 20/20 | | 17.5 |
| E3 | Executive & Proposal Content | 17/20 | | 15 |
| E4 | Executive & Proposal Content | 18/20 | | 14 |
| E5 | Executive & Proposal Content | 20/20 | | 13.7 |
| E6 | Executive & Proposal Content | 20/20 | | 13.2 |
Category mean SOW & Requirements Decomposition 20Executive & Proposal Content 18.8
▶ Long-context retrieval & synthesis 1g 120 / 120 20 6/6 n=1 17.3 run 2026-09-28
| Test | Category | Score | | tok/s |
| L1 | Long-Context — Needle Retrieval | 20/20 | | 14.8 |
| L2 | Long-Context — Grounded QA with Distractors | 20/20 | | 12.3 |
| L3 | Long-Context — Ops Log Reasoning | 20/20 | | 8.5 |
| L4 | Long-Context — Faithful Summarization | 20/20 | | 9.8 |
| L5 | Long-Context — Instruction Retention | 20/20 | | 6.1 |
| L6 | Long-Context — Full-Haystack QA | 20/20 | | 3.7 |
Category mean Long-Context — Needle Retrieval 20Long-Context — Grounded QA with Distractors 20Long-Context — Ops Log Reasoning 20Long-Context — Faithful Summarization 20Long-Context — Instruction Retention 20Long-Context — Full-Haystack QA 20
▶ Long-context multi-needle (MRCR) 1g2 44.1 / 60 14.7 3/3 n=1 13.2 run 2026-09-21
| Test | Category | Score | | tok/s |
| L7 | Long-Context — Multi-Needle Retrieval (MRCR) | 20/20 | | 14 |
| L8 | Long-Context — Multi-Needle Retrieval (MRCR) | 20/20 | | 12.1 |
| L9 | Long-Context — Multi-Needle Retrieval (MRCR) | 4/20 | | 8.2 |
▶ Live one-shot builds (runtime-verified) 1h 76 / 80 19 4/4 n=1 17.1 run 2026-09-23
| Test | Category | Score | | tok/s |
| H1 | Web-Estate One-Shot Builds | 16/20 | | 21.5 |
| H2 | Web-Estate One-Shot Builds | 20/20 | | 16.4 |
| H3 | Web-Estate One-Shot Builds | 20/20 | | 14.7 |
| H4 | Web-Estate One-Shot Builds | 20/20 | | 15.6 |
▶ Production replay (real agent workload) 1i 177.6 / 240 14.8 12/12 n=1 13.2 run 2026-10-10
| Test | Category | Score | | tok/s |
| I1 | Nightly Fix Replay | 8/20 | | 12.7 |
| I2 | Nightly Fix Replay | 8/20 | | 12.3 |
| I3 | Nightly Fix Replay | 9/20 | | 8.7 |
| I4 | Nightly Fix Replay | 16/20 | | 12 |
| I5 | Nightly Fix Replay | 15/20 | | 11.5 |
| I6 | Nightly Fix Replay | 12/20 | | 11.8 |
| I7 | Audit Judgment | 15/20 | | 14.4 |
| I8 | Audit Judgment | 20/20 | | 14.3 |
| I9 | Cron Dependency Reasoning | 17/20 | | 12 |
| I10 | Cron Dependency Reasoning | 19/20 | | 12 |
| I11 | Curation Judgment | 20/20 | | 13.1 |
| I12 | Curation Judgment | 18/20 | | 13.7 |
Category mean Nightly Fix Replay 11.3Audit Judgment 17.5Cron Dependency Reasoning 18Curation Judgment 19
▶ Brand-constrained static build (negative-constraint adherence) 1j 72 / 100 14.4 5/5 n=1 18.4 run 2026-09-24
| Test | Category | Score | | tok/s |
| J1 | Brand-Constrained Static Build | 12/20 | | 20.3 |
| J3 | Brand-Constrained Static Build | 20/20 | | 18.5 |
| J4 | Brand-Constrained Static Build | 20/20 | | 17.2 |
| J5 | Brand-Constrained Static Build | 9/20 | | 18 |
| J6 | Brand-Constrained Static Build | 11/20 | | 17.8 |
Each test is LLM-judged 0–20 against a fixed rubric; suite max = tests × 20.
An ADJ chip marks a cell carrying a subjective
adjustment from the maintainer; the judged score is the one shown, is unchanged,
and remains what the axes, ranks and tie bands are computed from. Hover for the reason.
Rows with a ▶ expand to the per-test breakdown.
Where fewer tests ran than the suite total, unrun tests count as zero toward the axis score.
See methodology.