eval-harness

metric

model_false_dismiss

better
lower
kind
quality: measured against a reference or by a judge
decides
no goal
weighed by
no goal
recorded on
orrery_triage_v1_labelled

01what it measures, and how

false_dismiss of the model's own decision, before the fleet's gates.

A project explains its own metric at length with METRIC_DESCRIPTIONS = {"model_false_dismiss": {"summary": "…", "details": "…"}}.

02on orrery_triage_v1_labelled

gold/orrery_triage_v1_labelledtriage_stub_escalate_v1: 0triage_stub_escalate_v1 · rescored 3d3a6afd: 0triage_stub_escalate_v1 · rescored 5f2c057c: 0triage_stub_escalate_v1 · rescored 77adc97b: 0triage_stub_escalate_v1 · rescored cd35dc1f: 0triage_gpt_61_sol_v1 · repeats ×3: 0.108triage_gpt_61_sol_v1 · rescored 3d3a6afd, repeats ×3: 0.108triage_gpt_61_sol_v1 · rescored 5f2c057c, repeats ×3: 0.108triage_gpt_61_sol_v1 · rescored 77adc97b, repeats ×3: 0.108triage_gpt_61_sol_v1 · rescored cd35dc1f, repeats ×3: 0.108triage_gpt_61_sol_v1: 0.118triage_gpt_61_sol_v1 · rescored 3d3a6afd: 0.118triage_gpt_61_sol_v1 · rescored 5f2c057c: 0.118triage_gpt_61_sol_v1 · rescored 77adc97b: 0.118triage_gpt_61_sol_v1 · rescored cd35dc1f: 0.118triage_gpt_6_luna_v1: 0.147triage_gpt_6_luna_v1 · rescored 5f2c057c: 0.147triage_gpt_6_luna_v1 · rescored 77adc97b: 0.147triage_gpt_6_luna_v1 · rescored cd35dc1f: 0.147triage_local_dgx_phi4_14b_v1: 0.176triage_local_mac_phi4_14b_v1: 0.176triage_local_dgx_phi4_14b_v1 · rescored 3d3a6afd: 0.176triage_local_mac_phi4_14b_v1 · rescored 3d3a6afd: 0.176triage_local_dgx_phi4_14b_v1 · rescored 5f2c057c: 0.176triage_local_mac_phi4_14b_v1 · rescored 5f2c057c: 0.176triage_local_dgx_phi4_14b_v1 · rescored 77adc97b: 0.176triage_local_mac_phi4_14b_v1 · rescored 77adc97b: 0.176triage_local_dgx_phi4_14b_v1 · rescored cd35dc1f: 0.176triage_local_mac_phi4_14b_v1 · rescored cd35dc1f: 0.176triage_gpt_6_luna_v1 · repeats ×3: 0.186triage_gpt_6_luna_v1 · rescored 5f2c057c, repeats ×3: 0.186triage_gpt_6_luna_v1 · rescored 77adc97b, repeats ×3: 0.186triage_gpt_6_luna_v1 · rescored cd35dc1f, repeats ×3: 0.186triage_local_x3_phi4_14b_v1: 0.206triage_local_x3_phi4_14b_v1 · rescored 3d3a6afd: 0.206triage_local_x3_phi4_14b_v1 · rescored 5f2c057c: 0.206triage_local_x3_phi4_14b_v1 · rescored 77adc97b: 0.206triage_local_x3_phi4_14b_v1 · rescored cd35dc1f: 0.206triage_gpt_6_luna_v2: 0.235triage_gpt_6_luna_v2 · rescored 3d3a6afd: 0.235triage_gpt_6_luna_v2 · rescored 5f2c057c: 0.235triage_local_dgx_llama31_8b_v1: 0.265triage_local_mac_llama31_8b_v1: 0.265triage_local_x3_gpt_oss_20b_v1: 0.265triage_local_x3_llama31_8b_v1: 0.265triage_local_dgx_llama31_8b_v1 · rescored 3d3a6afd: 0.265triage_local_mac_llama31_8b_v1 · rescored 3d3a6afd: 0.265triage_local_x3_llama31_8b_v1 · rescored 3d3a6afd: 0.265triage_local_dgx_llama31_8b_v1 · rescored 5f2c057c: 0.265triage_local_mac_llama31_8b_v1 · rescored 5f2c057c: 0.265triage_local_x3_llama31_8b_v1 · rescored 5f2c057c: 0.265triage_local_dgx_llama31_8b_v1 · rescored 77adc97b: 0.265triage_local_mac_llama31_8b_v1 · rescored 77adc97b: 0.265triage_local_x3_llama31_8b_v1 · rescored 77adc97b: 0.265triage_local_dgx_llama31_8b_v1 · rescored cd35dc1f: 0.265triage_local_mac_llama31_8b_v1 · rescored cd35dc1f: 0.265triage_local_x3_gpt_oss_20b_v1 · rescored cd35dc1f: 0.265triage_local_x3_llama31_8b_v1 · rescored cd35dc1f: 0.265triage_haiku_55_v1: 0.294triage_kimi_k31_pro_v1: 0.294triage_haiku_55_v1 · rescored 3d3a6afd: 0.294triage_haiku_55_v1 · rescored 5f2c057c: 0.294triage_kimi_k31_pro_v1 · rescored 5f2c057c: 0.294triage_haiku_55_v1 · rescored 77adc97b: 0.294triage_kimi_k31_pro_v1 · rescored 77adc97b: 0.294triage_haiku_55_v1 · rescored cd35dc1f: 0.294triage_kimi_k31_pro_v1 · rescored cd35dc1f: 0.294triage_haiku_55_v1 · repeats ×3: 0.304triage_haiku_55_v1 · rescored 3d3a6afd, repeats ×3: 0.304triage_haiku_55_v1 · rescored 5f2c057c, repeats ×3: 0.304triage_haiku_55_v1 · rescored 77adc97b, repeats ×3: 0.304triage_haiku_55_v1 · rescored cd35dc1f, repeats ×3: 0.304triage_local_x3_nemotron_35_bolt_e4b_v1: 0.324triage_local_x3_nemotron_35_lightning_30b_v1: 0.324triage_local_x3_nemotron_35_bolt_e4b_v1 · repeats: 0.324triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c: 0.324triage_local_x3_nemotron_35_lightning_30b_v1 · rescored 5f2c057c: 0.324triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c, repeats: 0.324triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 77adc97b: 0.324triage_local_x3_nemotron_35_lightning_30b_v1 · rescored 77adc97b: 0.324triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 77adc97b, repeats: 0.324triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f: 0.324triage_local_x3_nemotron_35_lightning_30b_v1 · rescored cd35dc1f: 0.324triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f, repeats: 0.324triage_local_dgx_nemotron_35_lightning_30b_v1: 0.353triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1: 0.353triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 3d3a6afd: 0.353triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored 5f2c057c: 0.353triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 5f2c057c: 0.353triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored 77adc97b: 0.353triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 77adc97b: 0.353triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored cd35dc1f: 0.353triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored cd35dc1f: 0.353triage_local_dgx_gpt_oss_20b_v1: 0.382triage_local_dgx_gpt_oss_20b_v1 · rescored cd35dc1f: 0.382triage_gemini_4_argon_v1: 0.412triage_local_dgx_nemotron_35_bolt_e4b_v1: 0.412triage_local_mac_nemotron_35_bolt_e4b_v1: 0.412triage_local_mac_nemotron_35_lightning_30b_v2: 0.412triage_local_x3_nemotron_35_lightning_30b_v2: 0.412triage_gemini_4_argon_v1 · rescored 3d3a6afd: 0.412triage_local_mac_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd: 0.412triage_local_x3_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd: 0.412triage_gemini_4_argon_v1 · rescored 5f2c057c: 0.412triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c: 0.412triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c: 0.412triage_local_mac_nemotron_35_lightning_30b_v2 · rescored 5f2c057c: 0.412triage_local_x3_nemotron_35_lightning_30b_v2 · rescored 5f2c057c: 0.412triage_gemini_4_argon_v1 · rescored 77adc97b: 0.412triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored 77adc97b: 0.412triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored 77adc97b: 0.412triage_gemini_4_argon_v1 · rescored cd35dc1f: 0.412triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f: 0.412triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f: 0.412triage_local_dgx_llama4_scout_v1: 0.441triage_local_dgx_nemotron_35_lightning_30b_v2: 0.441triage_local_mac_gpt_oss_20b_v1: 0.441triage_local_dgx_llama4_scout_v1 · rescored 3d3a6afd: 0.441triage_local_dgx_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd: 0.441triage_local_dgx_llama4_scout_v1 · rescored 5f2c057c: 0.441triage_local_dgx_nemotron_35_lightning_30b_v2 · rescored 5f2c057c: 0.441triage_local_dgx_llama4_scout_v1 · rescored 77adc97b: 0.441triage_local_dgx_llama4_scout_v1 · rescored cd35dc1f: 0.441triage_local_mac_gpt_oss_20b_v1 · rescored cd35dc1f: 0.441triage_local_dgx_gpt_oss_20b_v2: 0.471triage_local_dgx_nemotron_35_super_120b_v2: 0.471triage_local_mac_nemotron_35_lightning_30b_v1: 0.471triage_local_dgx_gpt_oss_20b_v2 · rescored 3d3a6afd: 0.471triage_local_dgx_nemotron_35_super_120b_v2 · rescored 3d3a6afd: 0.471triage_local_dgx_gpt_oss_20b_v2 · rescored 5f2c057c: 0.471triage_local_dgx_nemotron_35_super_120b_v2 · rescored 5f2c057c: 0.471triage_local_mac_nemotron_35_lightning_30b_v1 · rescored 5f2c057c: 0.471triage_local_mac_nemotron_35_lightning_30b_v1 · rescored 77adc97b: 0.471triage_local_mac_nemotron_35_lightning_30b_v1 · rescored cd35dc1f: 0.471triage_kimi_k31_pro_v2: 0.5triage_local_dgx_gpt_oss_120b_v1: 0.5triage_local_dgx_nemotron_35_super_120b_v1: 0.5triage_local_mac_gpt_oss_20b_v2: 0.5triage_kimi_k31_pro_v2 · rescored 3d3a6afd: 0.5triage_local_mac_gpt_oss_20b_v2 · rescored 3d3a6afd: 0.5triage_kimi_k31_pro_v2 · rescored 5f2c057c: 0.5triage_local_dgx_gpt_oss_120b_v1 · rescored 5f2c057c: 0.5triage_local_dgx_nemotron_35_super_120b_v1 · rescored 5f2c057c: 0.5triage_local_mac_gpt_oss_20b_v2 · rescored 5f2c057c: 0.5triage_local_dgx_gpt_oss_120b_v1 · rescored 77adc97b: 0.5triage_local_dgx_nemotron_35_super_120b_v1 · rescored 77adc97b: 0.5triage_local_dgx_gpt_oss_120b_v1 · rescored cd35dc1f: 0.5triage_local_dgx_nemotron_35_super_120b_v1 · rescored cd35dc1f: 0.5triage_baseline_pro_v1: 0.529triage_hosted_qwen3_30b_a3b_v1: 0.529triage_local_mac_nemotron_35_bolt_e4b_v2: 0.529triage_sonnet_55_v1: 0.529triage_baseline_pro_v1 · rescored 3d3a6afd: 0.529triage_hosted_qwen3_30b_a3b_v1 · rescored 3d3a6afd: 0.529triage_local_mac_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd: 0.529triage_sonnet_55_v1 · rescored 3d3a6afd: 0.529triage_baseline_pro_v1 · rescored 5f2c057c: 0.529triage_hosted_qwen3_30b_a3b_v1 · rescored 5f2c057c: 0.529triage_local_mac_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c: 0.529triage_local_x3_gpt_oss_20b_v1 · rescored 5f2c057c: 0.529triage_sonnet_55_v1 · rescored 5f2c057c: 0.529triage_baseline_pro_v1 · rescored 77adc97b: 0.529triage_hosted_qwen3_30b_a3b_v1 · rescored 77adc97b: 0.529triage_local_x3_gpt_oss_20b_v1 · rescored 77adc97b: 0.529triage_sonnet_55_v1 · rescored 77adc97b: 0.529triage_baseline_pro_v1 · rescored cd35dc1f: 0.529triage_hosted_qwen3_30b_a3b_v1 · rescored cd35dc1f: 0.529triage_sonnet_55_v1 · rescored cd35dc1f: 0.529triage_hosted_gpt_oss_120b_v1: 0.559triage_hosted_mistral_medium_35_v1: 0.559triage_local_dgx_gemma4_12b_noreason_v1: 0.559triage_local_dgx_gemma4_31b_v1: 0.559triage_local_dgx_mistral_medium_35_128b_v1: 0.559triage_local_dgx_nemotron_35_bolt_e4b_v2: 0.559triage_local_mac_gemma4_31b_v1: 0.559triage_hosted_mistral_medium_35_v1 · rescored 3d3a6afd: 0.559triage_local_dgx_gemma4_12b_noreason_v1 · rescored 3d3a6afd: 0.559triage_local_dgx_mistral_medium_35_128b_v1 · rescored 3d3a6afd: 0.559triage_local_dgx_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd: 0.559triage_hosted_gpt_oss_120b_v1 · rescored 5f2c057c: 0.559triage_hosted_mistral_medium_35_v1 · rescored 5f2c057c: 0.559triage_local_dgx_gemma4_12b_noreason_v1 · rescored 5f2c057c: 0.559triage_local_dgx_gemma4_31b_v1 · rescored 5f2c057c: 0.559triage_local_dgx_mistral_medium_35_128b_v1 · rescored 5f2c057c: 0.559triage_local_dgx_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c: 0.559triage_local_mac_gemma4_31b_v1 · rescored 5f2c057c: 0.559triage_hosted_gpt_oss_120b_v1 · rescored 77adc97b: 0.559triage_hosted_mistral_medium_35_v1 · rescored 77adc97b: 0.559triage_local_dgx_gemma4_12b_noreason_v1 · rescored 77adc97b: 0.559triage_local_dgx_gemma4_31b_v1 · rescored 77adc97b: 0.559triage_local_dgx_mistral_medium_35_128b_v1 · rescored 77adc97b: 0.559triage_local_mac_gemma4_31b_v1 · rescored 77adc97b: 0.559triage_hosted_gpt_oss_120b_v1 · rescored cd35dc1f: 0.559triage_hosted_mistral_medium_35_v1 · rescored cd35dc1f: 0.559triage_local_dgx_gemma4_12b_noreason_v1 · rescored cd35dc1f: 0.559triage_local_dgx_gemma4_31b_v1 · rescored cd35dc1f: 0.559triage_local_dgx_mistral_medium_35_128b_v1 · rescored cd35dc1f: 0.559triage_local_mac_gemma4_31b_v1 · rescored cd35dc1f: 0.559triage_hosted_gpt_oss_120b_v2: 0.588triage_hosted_qwen35_122b_v1: 0.588triage_local_dgx_gpt_oss_120b_v2: 0.588triage_local_mac_gemma4_12b_noreason_v1: 0.588triage_local_x3_gemma4_31b_v1: 0.588triage_local_x3_gpt_oss_20b_v2: 0.588triage_local_x3_nemotron_35_bolt_e4b_v2: 0.588triage_hosted_gpt_oss_120b_v2 · rescored 3d3a6afd: 0.588triage_local_dgx_gpt_oss_120b_v2 · rescored 3d3a6afd: 0.588triage_local_mac_gemma4_12b_noreason_v1 · rescored 3d3a6afd: 0.588triage_local_x3_gpt_oss_20b_v2 · rescored 3d3a6afd: 0.588triage_local_x3_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd: 0.588triage_hosted_gpt_oss_120b_v2 · rescored 5f2c057c: 0.588triage_hosted_qwen35_122b_v1 · rescored 5f2c057c: 0.588triage_local_dgx_gpt_oss_120b_v2 · rescored 5f2c057c: 0.588triage_local_dgx_gpt_oss_20b_v1 · rescored 5f2c057c: 0.588triage_local_mac_gemma4_12b_noreason_v1 · rescored 5f2c057c: 0.588triage_local_x3_gemma4_31b_v1 · rescored 5f2c057c: 0.588triage_local_x3_gpt_oss_20b_v2 · rescored 5f2c057c: 0.588triage_local_x3_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c: 0.588triage_hosted_qwen35_122b_v1 · rescored 77adc97b: 0.588triage_local_dgx_gpt_oss_20b_v1 · rescored 77adc97b: 0.588triage_local_mac_gemma4_12b_noreason_v1 · rescored 77adc97b: 0.588triage_local_x3_gemma4_31b_v1 · rescored 77adc97b: 0.588triage_hosted_qwen35_122b_v1 · rescored cd35dc1f: 0.588triage_local_mac_gemma4_12b_noreason_v1 · rescored cd35dc1f: 0.588triage_local_x3_gemma4_31b_v1 · rescored cd35dc1f: 0.588triage_baseline_pro_v1 · repeats ×3: 0.608triage_baseline_pro_v1 · rescored 3d3a6afd, repeats ×3: 0.608triage_baseline_pro_v1 · rescored 5f2c057c, repeats ×3: 0.608triage_baseline_pro_v1 · rescored 77adc97b, repeats ×3: 0.608triage_baseline_pro_v1 · rescored cd35dc1f, repeats ×3: 0.608triage_local_dgx_qwen35_122b_v1: 0.647triage_local_dgx_qwen35_122b_v1 · rescored 5f2c057c: 0.647triage_local_mac_gpt_oss_20b_v1 · rescored 5f2c057c: 0.647triage_local_dgx_qwen35_122b_v1 · rescored 77adc97b: 0.647triage_local_mac_gpt_oss_20b_v1 · rescored 77adc97b: 0.647triage_local_dgx_qwen35_122b_v1 · rescored cd35dc1f: 0.647triage_local_mac_qwen3_30b_a3b_v1: 0.676triage_local_x3_qwen3_30b_a3b_v1: 0.676triage_mistral_large_4_v1: 0.676triage_mistral_large_4_v2: 0.676triage_local_mac_qwen3_30b_a3b_v1 · rescored 3d3a6afd: 0.676triage_local_x3_qwen3_30b_a3b_v1 · rescored 3d3a6afd: 0.676triage_mistral_large_4_v2 · rescored 3d3a6afd: 0.676triage_local_mac_qwen3_30b_a3b_v1 · rescored 5f2c057c: 0.676triage_local_x3_qwen3_30b_a3b_v1 · rescored 5f2c057c: 0.676triage_mistral_large_4_v1 · rescored 5f2c057c: 0.676triage_mistral_large_4_v2 · rescored 5f2c057c: 0.676triage_local_mac_qwen3_30b_a3b_v1 · rescored 77adc97b: 0.676triage_local_x3_qwen3_30b_a3b_v1 · rescored 77adc97b: 0.676triage_mistral_large_4_v1 · rescored 77adc97b: 0.676triage_local_mac_qwen3_30b_a3b_v1 · rescored cd35dc1f: 0.676triage_local_x3_qwen3_30b_a3b_v1 · rescored cd35dc1f: 0.676triage_mistral_large_4_v1 · rescored cd35dc1f: 0.676triage_gemini_35_flash_lite_v1: 0.735triage_hosted_qwen35_122b_v2: 0.735triage_gemini_35_flash_lite_v1 · rescored 3d3a6afd: 0.735triage_hosted_qwen35_122b_v2 · rescored 3d3a6afd: 0.735triage_gemini_35_flash_lite_v1 · rescored 5f2c057c: 0.735triage_hosted_qwen35_122b_v2 · rescored 5f2c057c: 0.735triage_gemini_35_flash_lite_v1 · rescored 77adc97b: 0.735triage_gemini_35_flash_lite_v1 · rescored cd35dc1f: 0.735triage_local_dgx_qwen3_30b_a3b_v1: 0.765triage_local_dgx_qwen3_30b_a3b_v1 · rescored 3d3a6afd: 0.765triage_local_dgx_qwen3_30b_a3b_v1 · rescored 5f2c057c: 0.765triage_local_dgx_qwen3_30b_a3b_v1 · rescored 77adc97b: 0.765triage_local_dgx_qwen3_30b_a3b_v1 · rescored cd35dc1f: 0.765triage_hosted_gemma4_31b_v1: 0.794triage_local_x3_qwen38_27b_v2: 0.794triage_local_x3_qwen38_27b_v2 · rescored 3d3a6afd: 0.794triage_hosted_gemma4_31b_v1 · rescored 5f2c057c: 0.794triage_local_x3_qwen38_27b_v2 · rescored 5f2c057c: 0.794triage_hosted_gemma4_31b_v1 · rescored 77adc97b: 0.794triage_hosted_gemma4_31b_v1 · rescored cd35dc1f: 0.794triage_deepseek_41flash_v1: 0.824triage_local_dgx_qwen38_27b_v2: 0.824triage_local_mac_qwen38_27b_v1: 0.824triage_local_mac_qwen38_27b_v2: 0.824triage_local_x3_qwen38_27b_v1: 0.824triage_local_dgx_qwen38_27b_v2 · rescored 3d3a6afd: 0.824triage_local_mac_qwen38_27b_v2 · rescored 3d3a6afd: 0.824triage_deepseek_41flash_v1 · rescored 5f2c057c: 0.824triage_local_dgx_qwen38_27b_v2 · rescored 5f2c057c: 0.824triage_local_mac_qwen38_27b_v1 · rescored 5f2c057c: 0.824triage_local_mac_qwen38_27b_v2 · rescored 5f2c057c: 0.824triage_local_x3_qwen38_27b_v1 · rescored 5f2c057c: 0.824triage_deepseek_41flash_v1 · rescored 77adc97b: 0.824triage_local_mac_qwen38_27b_v1 · rescored 77adc97b: 0.824triage_local_x3_qwen38_27b_v1 · rescored 77adc97b: 0.824triage_deepseek_41flash_v1 · rescored cd35dc1f: 0.824triage_local_mac_qwen38_27b_v1 · rescored cd35dc1f: 0.824triage_local_x3_qwen38_27b_v1 · rescored cd35dc1f: 0.824triage_baseline_pro_noreason_v1: 0.853triage_deepseek_41flash_v2: 0.853triage_local_dgx_gemma4_31b_v2: 0.853triage_local_dgx_ministral_3_14b_v1: 0.853triage_local_dgx_qwen38_27b_v1: 0.853triage_baseline_pro_noreason_v1 · rescored 3d3a6afd: 0.853triage_deepseek_41flash_v2 · rescored 3d3a6afd: 0.853triage_local_dgx_gemma4_31b_v2 · rescored 3d3a6afd: 0.853triage_local_dgx_ministral_3_14b_v1 · rescored 3d3a6afd: 0.853triage_baseline_pro_noreason_v1 · rescored 5f2c057c: 0.853triage_deepseek_41flash_v2 · rescored 5f2c057c: 0.853triage_local_dgx_gemma4_31b_v2 · rescored 5f2c057c: 0.853triage_local_dgx_ministral_3_14b_v1 · rescored 5f2c057c: 0.853triage_local_dgx_qwen38_27b_v1 · rescored 5f2c057c: 0.853triage_local_dgx_ministral_3_14b_v1 · rescored 77adc97b: 0.853triage_local_dgx_qwen38_27b_v1 · rescored 77adc97b: 0.853triage_local_dgx_ministral_3_14b_v1 · rescored cd35dc1f: 0.853triage_local_dgx_qwen38_27b_v1 · rescored cd35dc1f: 0.853triage_hosted_gemma4_31b_v2: 0.882triage_local_dgx_qwen35_122b_v2: 0.882triage_local_mac_gemma4_31b_v2: 0.882triage_local_x3_gemma4_31b_v2: 0.882triage_hosted_gemma4_31b_v2 · rescored 3d3a6afd: 0.882triage_local_dgx_qwen35_122b_v2 · rescored 3d3a6afd: 0.882triage_local_mac_gemma4_31b_v2 · rescored 3d3a6afd: 0.882triage_local_x3_gemma4_31b_v2 · rescored 3d3a6afd: 0.882triage_hosted_gemma4_31b_v2 · rescored 5f2c057c: 0.882triage_local_dgx_qwen35_122b_v2 · rescored 5f2c057c: 0.882triage_local_mac_gemma4_31b_v2 · rescored 5f2c057c: 0.882triage_local_x3_gemma4_31b_v2 · rescored 5f2c057c: 0.882triage_local_mac_ministral_3_14b_v1: 0.912triage_local_mac_ministral_3_14b_v1 · rescored cd35dc1f: 0.912triage_local_x3_ministral_3_14b_v1: 0.941triage_local_mac_ministral_3_14b_v1 · rescored 3d3a6afd: 0.941triage_local_x3_ministral_3_14b_v1 · rescored 3d3a6afd: 0.941triage_local_mac_ministral_3_14b_v1 · rescored 5f2c057c: 0.941triage_local_x3_ministral_3_14b_v1 · rescored 5f2c057c: 0.941triage_local_mac_ministral_3_14b_v1 · rescored 77adc97b: 0.941triage_local_x3_ministral_3_14b_v1 · rescored 77adc97b: 0.941triage_local_x3_ministral_3_14b_v1 · rescored cd35dc1f: 0.941triage_stub_dismiss_v1: 1triage_stub_dismiss_v1 · rescored 3d3a6afd: 1triage_stub_dismiss_v1 · rescored 5f2c057c: 1triage_stub_dismiss_v1 · rescored 77adc97b: 1triage_stub_dismiss_v1 · rescored cd35dc1f: 10 · better1

scored against silverscored by a judgewrote the reference it is scored againstscored against a label set, not a reference setunmarked rows are scored against gold

armmean model_false_dismissitemsscored against
triage_stub_escalate_v10.00034gold/orrery_triage_v1_labelled gold
triage_stub_escalate_v1 · rescored 3d3a6afd0.00034gold/orrery_triage_v1_labelled gold
triage_stub_escalate_v1 · rescored 5f2c057c0.00034gold/orrery_triage_v1_labelled gold
triage_stub_escalate_v1 · rescored 77adc97b0.00034gold/orrery_triage_v1_labelled gold
triage_stub_escalate_v1 · rescored cd35dc1f0.00034gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · repeats ×30.10834gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 3d3a6afd, repeats ×30.10834gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 5f2c057c, repeats ×30.10834gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 77adc97b, repeats ×30.10834gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored cd35dc1f, repeats ×30.10834gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v10.11834gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 3d3a6afd0.11834gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 5f2c057c0.11834gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 77adc97b0.11834gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored cd35dc1f0.11834gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v10.14734gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored 5f2c057c0.14734gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored 77adc97b0.14734gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored cd35dc1f0.14734gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v10.17634gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v10.17634gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v1 · rescored 3d3a6afd0.17634gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v1 · rescored 3d3a6afd0.17634gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v1 · rescored 5f2c057c0.17634gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v1 · rescored 5f2c057c0.17634gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v1 · rescored 77adc97b0.17634gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v1 · rescored 77adc97b0.17634gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v1 · rescored cd35dc1f0.17634gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v1 · rescored cd35dc1f0.17634gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · repeats ×30.18634gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored 5f2c057c, repeats ×30.18634gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored 77adc97b, repeats ×30.18634gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored cd35dc1f, repeats ×30.18634gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v10.20634gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v1 · rescored 3d3a6afd0.20634gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v1 · rescored 5f2c057c0.20634gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v1 · rescored 77adc97b0.20634gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v1 · rescored cd35dc1f0.20634gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v20.23534gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v2 · rescored 3d3a6afd0.23534gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v2 · rescored 5f2c057c0.23534gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v10.26534gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v10.26534gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v10.26534gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v10.26534gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v1 · rescored 3d3a6afd0.26534gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v1 · rescored 3d3a6afd0.26534gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v1 · rescored 3d3a6afd0.26534gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v1 · rescored 5f2c057c0.26534gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v1 · rescored 5f2c057c0.26534gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v1 · rescored 5f2c057c0.26534gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v1 · rescored 77adc97b0.26534gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v1 · rescored 77adc97b0.26534gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v1 · rescored 77adc97b0.26534gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v1 · rescored cd35dc1f0.26534gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v1 · rescored cd35dc1f0.26534gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v1 · rescored cd35dc1f0.26534gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v1 · rescored cd35dc1f0.26534gold/orrery_triage_v1_labelled gold
triage_haiku_55_v10.29434gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v10.29434gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 3d3a6afd0.29434gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 5f2c057c0.29434gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v1 · rescored 5f2c057c0.29434gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 77adc97b0.29434gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v1 · rescored 77adc97b0.29434gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored cd35dc1f0.29434gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v1 · rescored cd35dc1f0.29434gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · repeats ×30.30434gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 3d3a6afd, repeats ×30.30434gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 5f2c057c, repeats ×30.30434gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 77adc97b, repeats ×30.30434gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored cd35dc1f, repeats ×30.30434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v10.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v10.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · repeats0.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c0.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v1 · rescored 5f2c057c0.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c, repeats0.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 77adc97b0.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v1 · rescored 77adc97b0.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 77adc97b, repeats0.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f0.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v1 · rescored cd35dc1f0.32434gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f, repeats0.32434gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v10.35334gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v10.35334gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 3d3a6afd0.35334gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored 5f2c057c0.35334gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 5f2c057c0.35334gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored 77adc97b0.35334gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 77adc97b0.35334gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored cd35dc1f0.35334gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored cd35dc1f0.35334gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v10.38234gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v1 · rescored cd35dc1f0.38234gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v10.41234gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v10.41234gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v10.41234gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v20.41234gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v20.41234gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v1 · rescored 3d3a6afd0.41234gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd0.41234gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd0.41234gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v1 · rescored 5f2c057c0.41234gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c0.41234gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c0.41234gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v2 · rescored 5f2c057c0.41234gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v2 · rescored 5f2c057c0.41234gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v1 · rescored 77adc97b0.41234gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored 77adc97b0.41234gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored 77adc97b0.41234gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v1 · rescored cd35dc1f0.41234gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f0.41234gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f0.41234gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v10.44134gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v20.44134gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v10.44134gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v1 · rescored 3d3a6afd0.44134gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd0.44134gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v1 · rescored 5f2c057c0.44134gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v2 · rescored 5f2c057c0.44134gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v1 · rescored 77adc97b0.44134gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v1 · rescored cd35dc1f0.44134gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v1 · rescored cd35dc1f0.44134gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v20.47134gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v20.47134gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v10.47134gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v2 · rescored 3d3a6afd0.47134gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v2 · rescored 3d3a6afd0.47134gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v2 · rescored 5f2c057c0.47134gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v2 · rescored 5f2c057c0.47134gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v1 · rescored 5f2c057c0.47134gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v1 · rescored 77adc97b0.47134gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v1 · rescored cd35dc1f0.47134gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v20.50034gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v10.50034gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v10.50034gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v20.50034gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v2 · rescored 3d3a6afd0.50034gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v2 · rescored 3d3a6afd0.50034gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v2 · rescored 5f2c057c0.50034gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v1 · rescored 5f2c057c0.50034gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v1 · rescored 5f2c057c0.50034gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v2 · rescored 5f2c057c0.50034gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v1 · rescored 77adc97b0.50034gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v1 · rescored 77adc97b0.50034gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v1 · rescored cd35dc1f0.50034gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v1 · rescored cd35dc1f0.50034gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v10.52934gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v10.52934gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v20.52934gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v10.52934gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 3d3a6afd0.52934gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v1 · rescored 3d3a6afd0.52934gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd0.52934gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v1 · rescored 3d3a6afd0.52934gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 5f2c057c0.52934gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v1 · rescored 5f2c057c0.52934gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c0.52934gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v1 · rescored 5f2c057c0.52934gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v1 · rescored 5f2c057c0.52934gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 77adc97b0.52934gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v1 · rescored 77adc97b0.52934gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v1 · rescored 77adc97b0.52934gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v1 · rescored 77adc97b0.52934gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored cd35dc1f0.52934gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v1 · rescored cd35dc1f0.52934gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v1 · rescored cd35dc1f0.52934gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v10.55934gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v10.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v10.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v10.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v10.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v20.55934gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v10.55934gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v1 · rescored 3d3a6afd0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v1 · rescored 3d3a6afd0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v1 · rescored 3d3a6afd0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd0.55934gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v1 · rescored 5f2c057c0.55934gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v1 · rescored 5f2c057c0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v1 · rescored 5f2c057c0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v1 · rescored 5f2c057c0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v1 · rescored 5f2c057c0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c0.55934gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v1 · rescored 5f2c057c0.55934gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v1 · rescored 77adc97b0.55934gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v1 · rescored 77adc97b0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v1 · rescored 77adc97b0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v1 · rescored 77adc97b0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v1 · rescored 77adc97b0.55934gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v1 · rescored 77adc97b0.55934gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v1 · rescored cd35dc1f0.55934gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v1 · rescored cd35dc1f0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v1 · rescored cd35dc1f0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v1 · rescored cd35dc1f0.55934gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v1 · rescored cd35dc1f0.55934gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v1 · rescored cd35dc1f0.55934gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v20.58834gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v10.58834gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v20.58834gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v10.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v10.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v20.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v20.58834gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v2 · rescored 3d3a6afd0.58834gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v2 · rescored 3d3a6afd0.58834gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v1 · rescored 3d3a6afd0.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v2 · rescored 3d3a6afd0.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd0.58834gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v2 · rescored 5f2c057c0.58834gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v1 · rescored 5f2c057c0.58834gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v2 · rescored 5f2c057c0.58834gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v1 · rescored 5f2c057c0.58834gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v1 · rescored 5f2c057c0.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v1 · rescored 5f2c057c0.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v2 · rescored 5f2c057c0.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c0.58834gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v1 · rescored 77adc97b0.58834gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v1 · rescored 77adc97b0.58834gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v1 · rescored 77adc97b0.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v1 · rescored 77adc97b0.58834gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v1 · rescored cd35dc1f0.58834gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v1 · rescored cd35dc1f0.58834gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v1 · rescored cd35dc1f0.58834gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · repeats ×30.60834gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 3d3a6afd, repeats ×30.60834gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 5f2c057c, repeats ×30.60834gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 77adc97b, repeats ×30.60834gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored cd35dc1f, repeats ×30.60834gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v10.64734gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v1 · rescored 5f2c057c0.64734gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v1 · rescored 5f2c057c0.64734gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v1 · rescored 77adc97b0.64734gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v1 · rescored 77adc97b0.64734gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v1 · rescored cd35dc1f0.64734gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v10.67634gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen3_30b_a3b_v10.67634gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v10.67634gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v20.67634gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v1 · rescored 3d3a6afd0.67634gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen3_30b_a3b_v1 · rescored 3d3a6afd0.67634gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v2 · rescored 3d3a6afd0.67634gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v1 · rescored 5f2c057c0.67634gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen3_30b_a3b_v1 · rescored 5f2c057c0.67634gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v1 · rescored 5f2c057c0.67634gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v2 · rescored 5f2c057c0.67634gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v1 · rescored 77adc97b0.67634gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen3_30b_a3b_v1 · rescored 77adc97b0.67634gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v1 · rescored 77adc97b0.67634gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v1 · rescored cd35dc1f0.67634gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen3_30b_a3b_v1 · rescored cd35dc1f0.67634gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v1 · rescored cd35dc1f0.67634gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v10.73534gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v20.73534gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v1 · rescored 3d3a6afd0.73534gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v2 · rescored 3d3a6afd0.73534gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v1 · rescored 5f2c057c0.73534gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v2 · rescored 5f2c057c0.73534gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v1 · rescored 77adc97b0.73534gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v1 · rescored cd35dc1f0.73534gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v10.76534gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v1 · rescored 3d3a6afd0.76534gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v1 · rescored 5f2c057c0.76534gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v1 · rescored 77adc97b0.76534gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v1 · rescored cd35dc1f0.76534gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v10.79434gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v20.79434gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v2 · rescored 3d3a6afd0.79434gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v1 · rescored 5f2c057c0.79434gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v2 · rescored 5f2c057c0.79434gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v1 · rescored 77adc97b0.79434gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v1 · rescored cd35dc1f0.79434gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v10.82434gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v20.82434gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v10.82434gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v20.82434gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v10.82434gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v2 · rescored 3d3a6afd0.82434gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v2 · rescored 3d3a6afd0.82434gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v1 · rescored 5f2c057c0.82434gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v2 · rescored 5f2c057c0.82434gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v1 · rescored 5f2c057c0.82434gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v2 · rescored 5f2c057c0.82434gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v1 · rescored 5f2c057c0.82434gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v1 · rescored 77adc97b0.82434gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v1 · rescored 77adc97b0.82434gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v1 · rescored 77adc97b0.82434gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v1 · rescored cd35dc1f0.82434gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v1 · rescored cd35dc1f0.82434gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v1 · rescored cd35dc1f0.82434gold/orrery_triage_v1_labelled gold
triage_baseline_pro_noreason_v10.85334gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v20.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v20.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v10.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v10.85334gold/orrery_triage_v1_labelled gold
triage_baseline_pro_noreason_v1 · rescored 3d3a6afd0.85334gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v2 · rescored 3d3a6afd0.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v2 · rescored 3d3a6afd0.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v1 · rescored 3d3a6afd0.85334gold/orrery_triage_v1_labelled gold
triage_baseline_pro_noreason_v1 · rescored 5f2c057c0.85334gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v2 · rescored 5f2c057c0.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v2 · rescored 5f2c057c0.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v1 · rescored 5f2c057c0.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v1 · rescored 5f2c057c0.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v1 · rescored 77adc97b0.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v1 · rescored 77adc97b0.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v1 · rescored cd35dc1f0.85334gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v1 · rescored cd35dc1f0.85334gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v20.88234gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v20.88234gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v20.88234gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v20.88234gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v2 · rescored 3d3a6afd0.88234gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v2 · rescored 3d3a6afd0.88234gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v2 · rescored 3d3a6afd0.88234gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v2 · rescored 3d3a6afd0.88234gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v2 · rescored 5f2c057c0.88234gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v2 · rescored 5f2c057c0.88234gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v2 · rescored 5f2c057c0.88234gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v2 · rescored 5f2c057c0.88234gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v10.91234gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v1 · rescored cd35dc1f0.91234gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v10.94134gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v1 · rescored 3d3a6afd0.94134gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v1 · rescored 3d3a6afd0.94134gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v1 · rescored 5f2c057c0.94134gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v1 · rescored 5f2c057c0.94134gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v1 · rescored 77adc97b0.94134gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v1 · rescored 77adc97b0.94134gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v1 · rescored cd35dc1f0.94134gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v11.00034gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v1 · rescored 3d3a6afd1.00034gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v1 · rescored 5f2c057c1.00034gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v1 · rescored 77adc97b1.00034gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v1 · rescored cd35dc1f1.00034gold/orrery_triage_v1_labelled gold