Codex CLI 0.104.0

GPT 5.3 Codex 2026-02-24
Score
21.73
PASSED
Pass Rate: 69.2%
go: 100.0% (6 tasks)
go
rust: 83.3% (6 tasks)
rus
typescript: 80.0% (5 tasks)
typ
dart: 33.3% (3 tasks)
dar
kotlin: 33.3% (3 tasks)
kot
zig: 33.3% (3 tasks)
zig

Evaluation Report

Summary

Metric Value
Agent codex
Model gpt-5.3-codex
Reasoning Effort low
Legacy Mode Yes
Timestamp 2026-02-24T122937
Pass Rate 69.2% (18/26)
Weighted Pass Rate 65.3%
Weighted Score 21.73 / 33.29
Duration 1077.8s

Quality Breakdown

  • Clean Passes: 18
  • Partial Passes (timed out but passed): 0
  • Integrity Violations (modified test files): 0
  • Failures: 8

Results by Language

Language Passed Failed Total Pass Rate
dart 1 2 3 33.3%
go 6 0 6 100.0%
kotlin 1 2 3 33.3%
rust 5 1 6 83.3%
typescript 4 1 5 80.0%
zig 1 2 3 33.3%

Results by Tier

Tier Passed Failed Total Pass Rate
core 10 2 12 83.3%
extended 8 6 14 57.1%

Task Results

Task Status Weight Score Duration
dart/future-pool ✘ FAIL 1.46 0.00 32.0s
dart/isolate-pool ✘ FAIL 1.50 0.00 87.9s
dart/reactive-cache ✔ PASS 1.50 1.50 48.6s
go/bank-account ✔ PASS 1.04 1.04 19.4s
go/dining-philosophers ✔ PASS 1.04 1.04 26.9s
go/errgroup-limit ✔ PASS 1.14 1.14 37.1s
go/parallel-letter-frequency ✔ PASS 1.04 1.04 25.9s
go/react ✔ PASS 1.14 1.14 57.4s
go/singleflight ✔ PASS 1.28 1.28 15.6s
kotlin/channel-multiplexer ✘ FAIL 1.50 0.00 72.5s
kotlin/flow-processor ✘ FAIL 1.50 0.00 113.4s
kotlin/lru-cache ✔ PASS 1.09 1.09 40.0s
rust/circular-buffer ✔ PASS 1.12 1.12 17.8s
rust/doubly-linked-list ✔ PASS 1.24 1.24 26.5s
rust/generational-arena ✔ PASS 1.24 1.24 21.2s
rust/macros ✔ PASS 1.50 1.50 18.1s
rust/parallel-letter-frequency ✔ PASS 1.12 1.12 16.6s
rust/regex-lite ✘ FAIL 1.40 0.00 19.9s
typescript/csv-lite ✔ PASS 1.36 1.36 30.0s
typescript/forth ✔ PASS 1.26 1.26 36.1s
typescript/glob ✔ PASS 1.14 1.14 22.0s
typescript/promise-pool ✘ FAIL 1.20 0.00 22.1s
typescript/react ✔ PASS 1.14 1.14 46.4s
zig/arena-allocator ✘ FAIL 1.50 0.00 90.4s
zig/comptime-json ✘ FAIL 1.50 0.00 101.1s
zig/small-vector ✔ PASS 1.34 1.34 32.6s

Verification

  • Harness Version: v1.6.1
  • Weight Version: 2.0
  • Tasks Hash: blake3:168696ded119da3ea973f15bde9c00151c127b3f54168196a84212827fa1b1f0
  • Results Hash: blake3:2cad03323a0b9fc54d2a15df4685a87e7c1a7df3643d204763d7ed6df219120b

Generated by SanityHarness on 2026-02-24T122937

Run Statistics

Tasks Passed 18 / 26
Duration 18m
Cost (Est) -
By Language
dart 33%
go 100%
kotlin 33%
rust 83%
typescript 80%
zig 33%