Codex CLI 0.104.0

GPT 5.2 Codex 2026-02-24
Score
17.39
PASSED
Pass Rate: 57.7%
go: 100.0% (6 tasks)
go
typescript: 80.0% (5 tasks)
typ
rust: 66.7% (6 tasks)
rus
kotlin: 33.3% (3 tasks)
kot
dart: 0.0% (3 tasks)
dar
zig: 0.0% (3 tasks)
zig

Evaluation Report

Summary

Metric Value
Agent codex
Model gpt-5.2-codex
Reasoning Effort medium
Legacy Mode Yes
Timestamp 2026-02-24T143559
Pass Rate 57.7% (15/26)
Weighted Pass Rate 52.2%
Weighted Score 17.39 / 33.29
Duration 1399.2s

Quality Breakdown

  • Clean Passes: 15
  • Partial Passes (timed out but passed): 0
  • Integrity Violations (modified test files): 0
  • Failures: 11

Results by Language

Language Passed Failed Total Pass Rate
dart 0 3 3 0.0%
go 6 0 6 100.0%
kotlin 1 2 3 33.3%
rust 4 2 6 66.7%
typescript 4 1 5 80.0%
zig 0 3 3 0.0%

Results by Tier

Tier Passed Failed Total Pass Rate
core 9 3 12 75.0%
extended 6 8 14 42.9%

Task Results

Task Status Weight Score Duration
dart/future-pool ✘ FAIL 1.46 0.00 52.3s
dart/isolate-pool ✘ FAIL 1.50 0.00 102.3s
dart/reactive-cache ✘ FAIL 1.50 0.00 56.2s
go/bank-account ✔ PASS 1.04 1.04 17.0s
go/dining-philosophers ✔ PASS 1.04 1.04 28.9s
go/errgroup-limit ✔ PASS 1.14 1.14 57.5s
go/parallel-letter-frequency ✔ PASS 1.04 1.04 38.7s
go/react ✔ PASS 1.14 1.14 97.5s
go/singleflight ✔ PASS 1.28 1.28 26.9s
kotlin/channel-multiplexer ✘ FAIL 1.50 0.00 90.3s
kotlin/flow-processor ✘ FAIL 1.50 0.00 92.2s
kotlin/lru-cache ✔ PASS 1.09 1.09 41.1s
rust/circular-buffer ✔ PASS 1.12 1.12 33.1s
rust/doubly-linked-list ✔ PASS 1.24 1.24 26.6s
rust/generational-arena ✔ PASS 1.24 1.24 27.5s
rust/macros ✘ FAIL 1.50 0.00 31.1s
rust/parallel-letter-frequency ✔ PASS 1.12 1.12 21.8s
rust/regex-lite ✘ FAIL 1.40 0.00 21.4s
typescript/csv-lite ✔ PASS 1.36 1.36 82.7s
typescript/forth ✔ PASS 1.26 1.26 46.4s
typescript/glob ✔ PASS 1.14 1.14 28.3s
typescript/promise-pool ✘ FAIL 1.20 0.00 25.0s
typescript/react ✔ PASS 1.14 1.14 50.4s
zig/arena-allocator ✘ FAIL 1.50 0.00 134.7s
zig/comptime-json ✘ FAIL 1.50 0.00 135.3s
zig/small-vector ✘ FAIL 1.34 0.00 34.0s

Verification

  • Harness Version: v1.6.1
  • Weight Version: 2.0
  • Tasks Hash: blake3:168696ded119da3ea973f15bde9c00151c127b3f54168196a84212827fa1b1f0
  • Results Hash: blake3:ebda120bcecfdcccab082391e05630cf275a664730229a662453a4335ad0d3ea

Generated by SanityHarness on 2026-02-24T143559

Run Statistics

Tasks Passed 15 / 26
Duration 23m
Cost (Est) -
By Language
dart 0%
go 100%
kotlin 33%
rust 67%
typescript 80%
zig 0%