Codex CLI 0.104.0

GPT 5.3 Codex 2026-02-24
Score
20.79
PASSED
Pass Rate: 69.2%
kotlin: 100.0% (3 tasks)
kot
go: 83.3% (6 tasks)
go
rust: 83.3% (6 tasks)
rus
typescript: 60.0% (5 tasks)
typ
dart: 33.3% (3 tasks)
dar
zig: 33.3% (3 tasks)
zig

Evaluation Report

Summary

Metric Value
Agent codex
Model gpt-5.3-codex
Reasoning Effort xhigh
Legacy Mode Yes
Timestamp 2026-02-24T124617
Pass Rate 69.2% (18/26)
Weighted Pass Rate 62.5%
Weighted Score 20.79 / 33.29
Duration 3132.9s

Quality Breakdown

  • Clean Passes: 14
  • Partial Passes (timed out but passed): 4
  • Integrity Violations (modified test files): 0
  • Failures: 4

Results by Language

Language Passed Failed Total Pass Rate
dart 1 2 3 33.3%
go 5 1 6 83.3%
kotlin 3 0 3 100.0%
rust 5 1 6 83.3%
typescript 3 2 5 60.0%
zig 1 2 3 33.3%

Results by Tier

Tier Passed Failed Total Pass Rate
core 8 4 12 66.7%
extended 10 4 14 71.4%

Task Results

Task Status Weight Score Duration
dart/future-pool ✘ FAIL 1.46 0.00 71.5s
dart/isolate-pool ✘ FAIL 1.50 0.00 196.6s
dart/reactive-cache ⚠ PARTIAL 1.50 1.12 257.5s
go/bank-account ✔ PASS 1.04 1.04 74.7s
go/dining-philosophers ✔ PASS 1.04 1.04 114.5s
go/errgroup-limit ✘ FAIL 1.14 0.00 120.9s
go/parallel-letter-frequency ✔ PASS 1.04 1.04 55.9s
go/react ✔ PASS 1.14 1.14 135.1s
go/singleflight ✔ PASS 1.28 1.28 61.1s
kotlin/channel-multiplexer ⚠ PARTIAL 1.50 1.12 251.8s
kotlin/flow-processor ✔ PASS 1.50 1.50 183.7s
kotlin/lru-cache ✔ PASS 1.09 1.09 53.6s
rust/circular-buffer ✔ PASS 1.12 1.12 47.9s
rust/doubly-linked-list ✔ PASS 1.24 1.24 110.2s
rust/generational-arena ✔ PASS 1.24 1.24 95.7s
rust/macros ⚠ PARTIAL 1.50 1.12 121.1s
rust/parallel-letter-frequency ✔ PASS 1.12 1.12 58.7s
rust/regex-lite ✘ FAIL 1.40 0.00 39.1s
typescript/csv-lite ✘ FAIL 1.36 0.00 121.0s
typescript/forth ⚠ PARTIAL 1.26 0.95 121.0s
typescript/glob ✔ PASS 1.14 1.14 52.4s
typescript/promise-pool ✘ FAIL 1.20 0.00 120.9s
typescript/react ✔ PASS 1.14 1.14 98.0s
zig/arena-allocator ✘ FAIL 1.50 0.00 245.6s
zig/comptime-json ✘ FAIL 1.50 0.00 244.7s
zig/small-vector ✔ PASS 1.34 1.34 80.0s

Verification

  • Harness Version: v1.6.1
  • Weight Version: 2.0
  • Tasks Hash: blake3:168696ded119da3ea973f15bde9c00151c127b3f54168196a84212827fa1b1f0
  • Results Hash: blake3:7e8a17fd2721df80f0b0fe0aecc2a6836ff85b7b416f6ab368ed1cfde3f4b618

Generated by SanityHarness on 2026-02-24T124617

Run Statistics

Tasks Passed 18 / 26
Duration 52m
Cost (Est) -
By Language
dart 33%
go 83%
kotlin 100%
rust 83%
typescript 60%
zig 33%