{% extends "_base.html" %} {% block title %}harness-evaluator — {{ run_name }}{% endblock %} {% block head_extra %} {% if is_in_progress %}{% endif %} {% endblock %} {% macro th_sort(col, label, sort, order, base_query, run_name) %} {{ label }}{% if sort == col %} {% endif %} {% endmacro %} {% block content %}

{{ run_name }}

{% if state_summary %}
{% for status, count in state_summary.items() %} {{ status }}: {{ count }} {% endfor %}
{% endif %} {% if failed_cells %}

Failed / Skipped Cells ({{ failed_cells|length }})

{% for c in failed_cells %} {% endfor %}
Cell IDStatusError
{{ c.cell_id }} {{ c.status }} {{ c.error or '' }}
{% endif %}
{{ total_cells }}
Total Cells
{{ passed }}
Passed
${{ "%.4f"|format(cost) }}
Cost
{{ total_count }}
Filtered
{% if run_metadata %}
Run Metadata
Created
{{ run_metadata.created_at }}
{% if run_metadata.harness_evaluator_version %}
Version
{{ run_metadata.harness_evaluator_version }}
{% endif %} {% if run_metadata.docker_image %}
Docker Image
{{ run_metadata.docker_image }}
{% endif %}
{% if run_metadata.config_json %}
{{ run_metadata.config_json }}
{% endif %}
{% endif %}
Export CSV Export JSON
{% if leaderboards %}

Leaderboards

{% for model_name, rows in leaderboards.items() %}

{{ model_name }}

{% for row in rows %} {% set sv = row.success_pct | float %} {% endfor %}
HarnessSuccess Avg TokensAvg Cost Avg TimeAPI Calls
{{ row.harness }} {{ row.success_pct }}% {{ row.avg_tokens }} ${{ row.avg_cost }} {{ row.avg_time_s }}s {{ row.avg_api_calls }}
{% endfor %}
{% endif %}

Results ({{ total_count }} total, showing {{ results|length }})

{{ th_sort('harness', 'Harness', sort, order, base_query, run_name) }} {{ th_sort('model', 'Model', sort, order, base_query, run_name) }} {{ th_sort('task_id', 'Task', sort, order, base_query, run_name) }} {{ th_sort('repeat', 'Repeat', sort, order, base_query, run_name) }} {{ th_sort('exit_class', 'Exit', sort, order, base_query, run_name) }} {{ th_sort('success', 'Success', sort, order, base_query, run_name) }} {{ th_sort('total_cost', 'Cost', sort, order, base_query, run_name) }} {{ th_sort('latency_ms', 'Time', sort, order, base_query, run_name) }} {{ th_sort('num_api_calls', 'API Calls', sort, order, base_query, run_name) }} {% for r in results %} {% set total_tokens = r.input_tokens + r.output_tokens + r.cache_read_tokens + r.cache_write_tokens + r.reasoning_tokens %} {% set exit_class = 'pass' if r.exit_class == 'pass' else ('partial' if r.exit_class == 'partial' else ('kill' if r.exit_class in ('retryable_kill', 'non_retryable_kill') else 'fail')) %} {% endfor %}
Error Class Error Message
{{ r.harness }} {{ r.model }} {{ r.task_id }} r{{ r.repeat }} {{ r.exit_class }} {{ "%.2f"|format(r.success) }} ${{ "%.6f"|format(r.total_cost) }} {{ "%.1f"|format(r.latency_ms / 1000) }}s {{ r.num_api_calls }} {{ r.error_class or '' }} {{ r.error_message or '' }}
{% if phase_results %}

Phase Details (current page)

{% for cell_id, phases in phase_results.items() %}
{{ cell_id }} ({{ phases|length }} phase{{ 's' if phases|length != 1 else '' }})
{% for p in phases %} {% endfor %}
PhaseModelRole Exit CodeDurationTimed Out TokensCostError
{{ p.phase_name }} {{ p.model }} {{ p.model_role }} {{ p.exit_code }} {{ "%.1f"|format(p.duration_ms / 1000) }}s {{ "yes" if p.timed_out else "no" }} {{ p.input_tokens + p.output_tokens }} ${{ "%.6f"|format(p.total_cost) }} {{ p.error or '' }}
{% endfor %} {% endif %} {% if total_pages > 1 %} {% endif %}
{% endblock %}