Quality Evaluation
Horizon RunMap completed and published an IFEval evaluation covering every one of its 541 prompts and 834 instructions. The published prompt-level loose result is 45.66%, with all responses available to inspect. The public app also completed the full IFEval process on another computer with an RTX 4060. Public app.
IFEval standalone benchmark
541 prompts and 834 instructions. Evidence level E1; experiment type standalone_benchmark; validation valid; publication public; independent reproduction not_attempted.
| Metric | Passed / total | Result |
|---|---|---|
| Prompt strict | 217 / 541 | 40.11% |
| Prompt loose | 247 / 541 | 45.66% |
| Instruction strict | 433 / 834 | 51.92% |
| Instruction loose | 474 / 834 | 56.83% |
The frozen evaluator and retained responses support CPU score recomputation. The published score belongs to this edition; the app's RTX 4060 execution is described as a separate completed milestone.
Explore all retained responses · Summary JSON · Experiment contract · Method · Verifier · Manifest · Evidence ZIP
Earlier paired BF16 / Horizon structural comparison
120 paired prompts. Evidence level E3; validation valid; publication public; independent reproduction not_attempted.
Frozen structural checks were satisfied by 29 BF16 responses and 28 Horizon responses. Horizon had 8 losses and 7 gains under those checks. These are structural categories, not answer-correctness grades.
CPU recomputation of frozen classifications from retained response pairs; no answer-correctness grade or generation rerun.
Paired results JSON · Original report · Method · Contract · Verifier · Manifest · Evidence ZIP