eval-harness

metric

correct

better
higher
kind
quality: measured against a reference or by a judge
decides
Encyclopedia topic classification (DBpedia), News topic classification (AG News)
weighed by
no goal
recorded on
ag_news_200, dbpedia_280, orrery_triage_v1_labelled

01what it measures, and how

1 when the decision is one the label accepts (file, dismiss, escalate or cleanup), else 0.

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

02on ag_news_200

gold/ag_news_200ag_bert_base_ta_n200_v1: 0.96ag_bert_base_ta_n200_v1 · linux: 0.96ag_bert_base_fy_n200_v1: 0.95ag_bert_base_fy_n200_v1 · linux: 0.95ag_bert_mini_n200_v1: 0.945ag_bert_mini_n200_v1 · linux: 0.945ag_anthropic_m_n200_v1: 0.91ag_anthropic_l_n200_v1: 0.9ag_openai_m_n200_v1: 0.895ag_openai_l_n200_v1: 0.89ag_qwen_m_n200_v1: 0.885ag_llama_m_n200_v1: 0.88ag_gemma_s_n200_v1: 0.875ag_mistral_l_n200_v1: 0.875ag_qwen_s_n200_v1: 0.875ag_deepseek_l_n200_v1: 0.87ag_gemma_l_n200_v1: 0.87ag_qwen_l_n200_v1: 0.87ag_glm_l_n200_v1: 0.86ag_glm_m_n200_v1: 0.86ag_mistral_m_n200_v1: 0.86ag_mistral_s_n200_v1: 0.86ag_openai_s_n200_v1: 0.86ag_gemma_m_n200_v1: 0.855ag_glm_s_n200_v1: 0.85ag_deepseek_m_n200_v1: 0.845ag_anthropic_s_n200_v1: 0.84ag_deepseek_s_n200_v1: 0.84ag_llama_l_n200_v1: 0.84ag_llama_s_n200_v1: 0.835ag_bart_mnli_n200_v1: 0.7ag_keyword_n200_v1: 0.67ag_constant_n200_v1: 0.25gold/ag_news_200_adj1ag_anthropic_l_n200_v1 · adjudicated: 0.985ag_openai_m_n200_v1 · adjudicated: 0.98ag_openai_l_n200_v1 · adjudicated: 0.975ag_anthropic_m_n200_v1 · adjudicated: 0.965ag_llama_m_n200_v1 · adjudicated: 0.965ag_gemma_s_n200_v1 · adjudicated: 0.96ag_qwen_m_n200_v1 · adjudicated: 0.96ag_qwen_s_n200_v1 · adjudicated: 0.96ag_deepseek_l_n200_v1 · adjudicated: 0.955ag_gemma_l_n200_v1 · adjudicated: 0.955ag_qwen_l_n200_v1 · adjudicated: 0.955ag_mistral_l_n200_v1 · adjudicated: 0.95ag_glm_l_n200_v1 · adjudicated: 0.945ag_glm_m_n200_v1 · adjudicated: 0.945ag_mistral_m_n200_v1 · adjudicated: 0.945ag_openai_s_n200_v1 · adjudicated: 0.945ag_gemma_m_n200_v1 · adjudicated: 0.94ag_glm_s_n200_v1 · adjudicated: 0.935ag_mistral_s_n200_v1 · adjudicated: 0.93ag_anthropic_s_n200_v1 · adjudicated: 0.925ag_deepseek_m_n200_v1 · adjudicated: 0.925ag_deepseek_s_n200_v1 · adjudicated: 0.925ag_llama_l_n200_v1 · adjudicated: 0.925ag_llama_s_n200_v1 · adjudicated: 0.92ag_bert_base_fy_n200_v1 · adjudicated: 0.905ag_bert_base_ta_n200_v1 · adjudicated: 0.905ag_bert_mini_n200_v1 · adjudicated: 0.905ag_bart_mnli_n200_v1 · adjudicated: 0.755ag_keyword_n200_v1 · adjudicated: 0.67ag_constant_n200_v1 · adjudicated: 0.2250.225better · 0.985

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 correctitemsscored against
ag_bert_base_ta_n200_v10.960200gold/ag_news_200 gold
ag_bert_base_ta_n200_v1 · linux0.960200gold/ag_news_200 gold
ag_bert_base_fy_n200_v10.950200gold/ag_news_200 gold
ag_bert_base_fy_n200_v1 · linux0.950200gold/ag_news_200 gold
ag_bert_mini_n200_v10.945200gold/ag_news_200 gold
ag_bert_mini_n200_v1 · linux0.945200gold/ag_news_200 gold
ag_anthropic_m_n200_v10.910200gold/ag_news_200 gold
ag_anthropic_l_n200_v10.900200gold/ag_news_200 gold
ag_openai_m_n200_v10.895200gold/ag_news_200 gold
ag_openai_l_n200_v10.890200gold/ag_news_200 gold
ag_qwen_m_n200_v10.885200gold/ag_news_200 gold
ag_llama_m_n200_v10.880200gold/ag_news_200 gold
ag_gemma_s_n200_v10.875200gold/ag_news_200 gold
ag_mistral_l_n200_v10.875200gold/ag_news_200 gold
ag_qwen_s_n200_v10.875200gold/ag_news_200 gold
ag_deepseek_l_n200_v10.870200gold/ag_news_200 gold
ag_gemma_l_n200_v10.870200gold/ag_news_200 gold
ag_qwen_l_n200_v10.870200gold/ag_news_200 gold
ag_glm_l_n200_v10.860200gold/ag_news_200 gold
ag_glm_m_n200_v10.860200gold/ag_news_200 gold
ag_mistral_m_n200_v10.860200gold/ag_news_200 gold
ag_mistral_s_n200_v10.860200gold/ag_news_200 gold
ag_openai_s_n200_v10.860200gold/ag_news_200 gold
ag_gemma_m_n200_v10.855200gold/ag_news_200 gold
ag_glm_s_n200_v10.850200gold/ag_news_200 gold
ag_deepseek_m_n200_v10.845200gold/ag_news_200 gold
ag_anthropic_s_n200_v10.840200gold/ag_news_200 gold
ag_deepseek_s_n200_v10.840200gold/ag_news_200 gold
ag_llama_l_n200_v10.840200gold/ag_news_200 gold
ag_llama_s_n200_v10.835200gold/ag_news_200 gold
ag_bart_mnli_n200_v10.700200gold/ag_news_200 gold
ag_keyword_n200_v10.670200gold/ag_news_200 gold
ag_constant_n200_v10.250200gold/ag_news_200 gold
ag_anthropic_l_n200_v1 · adjudicated0.985200gold/ag_news_200_adj1 gold
ag_openai_m_n200_v1 · adjudicated0.980200gold/ag_news_200_adj1 gold
ag_openai_l_n200_v1 · adjudicated0.975200gold/ag_news_200_adj1 gold
ag_anthropic_m_n200_v1 · adjudicated0.965200gold/ag_news_200_adj1 gold
ag_llama_m_n200_v1 · adjudicated0.965200gold/ag_news_200_adj1 gold
ag_gemma_s_n200_v1 · adjudicated0.960200gold/ag_news_200_adj1 gold
ag_qwen_m_n200_v1 · adjudicated0.960200gold/ag_news_200_adj1 gold
ag_qwen_s_n200_v1 · adjudicated0.960200gold/ag_news_200_adj1 gold
ag_deepseek_l_n200_v1 · adjudicated0.955200gold/ag_news_200_adj1 gold
ag_gemma_l_n200_v1 · adjudicated0.955200gold/ag_news_200_adj1 gold
ag_qwen_l_n200_v1 · adjudicated0.955200gold/ag_news_200_adj1 gold
ag_mistral_l_n200_v1 · adjudicated0.950200gold/ag_news_200_adj1 gold
ag_glm_l_n200_v1 · adjudicated0.945200gold/ag_news_200_adj1 gold
ag_glm_m_n200_v1 · adjudicated0.945200gold/ag_news_200_adj1 gold
ag_mistral_m_n200_v1 · adjudicated0.945200gold/ag_news_200_adj1 gold
ag_openai_s_n200_v1 · adjudicated0.945200gold/ag_news_200_adj1 gold
ag_gemma_m_n200_v1 · adjudicated0.940200gold/ag_news_200_adj1 gold
ag_glm_s_n200_v1 · adjudicated0.935200gold/ag_news_200_adj1 gold
ag_mistral_s_n200_v1 · adjudicated0.930200gold/ag_news_200_adj1 gold
ag_anthropic_s_n200_v1 · adjudicated0.925200gold/ag_news_200_adj1 gold
ag_deepseek_m_n200_v1 · adjudicated0.925200gold/ag_news_200_adj1 gold
ag_deepseek_s_n200_v1 · adjudicated0.925200gold/ag_news_200_adj1 gold
ag_llama_l_n200_v1 · adjudicated0.925200gold/ag_news_200_adj1 gold
ag_llama_s_n200_v1 · adjudicated0.920200gold/ag_news_200_adj1 gold
ag_bert_base_fy_n200_v1 · adjudicated0.905200gold/ag_news_200_adj1 gold
ag_bert_base_ta_n200_v1 · adjudicated0.905200gold/ag_news_200_adj1 gold
ag_bert_mini_n200_v1 · adjudicated0.905200gold/ag_news_200_adj1 gold
ag_bart_mnli_n200_v1 · adjudicated0.755200gold/ag_news_200_adj1 gold
ag_keyword_n200_v1 · adjudicated0.670200gold/ag_news_200_adj1 gold
ag_constant_n200_v1 · adjudicated0.225200gold/ag_news_200_adj1 gold

03on dbpedia_280

gold/dbpedia_280db_qwen_m_n200_v1: 0.993db_anthropic_l_n200_v1: 0.989db_bert_base_fy_n200_v1: 0.986db_deepseek_l_n200_v1: 0.986db_gemma_m_n200_v1: 0.986db_glm_m_n200_v1: 0.986db_qwen_s_n200_v1: 0.986db_bert_base_fy_n200_v1 · linux: 0.986db_anthropic_m_n200_v1: 0.982db_openai_m_n200_v1: 0.982db_anthropic_s_n200_v1: 0.979db_llama_m_n200_v1: 0.979db_deepseek_m_n200_v1: 0.975db_gemma_l_n200_v1: 0.975db_gemma_s_n200_v1: 0.975db_llama_l_n200_v1: 0.975db_openai_l_n200_v1: 0.975db_openai_s_n200_v1: 0.975db_glm_l_n200_v1: 0.968db_qwen_l_n200_v1: 0.968db_mistral_l_n200_v1: 0.964db_llama_s_n200_v1: 0.957db_mistral_m_n200_v1: 0.957db_mistral_s_n200_v1: 0.957db_deepseek_s_n200_v1: 0.943db_glm_s_n200_v1: 0.939db_keyword_n200_v1: 0.707db_bart_mnli_n200_v1: 0.629db_constant_n200_v1: 0.0710.071better · 0.993

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

04on orrery_triage_v1_labelled

gold/orrery_triage_v1_labelledtriage_local_x3_qwen3_30b_a3b_v1: 0.692triage_local_x3_qwen3_30b_a3b_v1 · rescored 3d3a6afd: 0.692triage_local_x3_qwen3_30b_a3b_v1 · rescored 5f2c057c: 0.692triage_local_x3_qwen3_30b_a3b_v1 · rescored 77adc97b: 0.692triage_local_x3_qwen3_30b_a3b_v1 · rescored cd35dc1f: 0.692triage_local_x3_nemotron_35_bolt_e4b_v2: 0.677triage_local_x3_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd: 0.677triage_local_x3_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c: 0.677triage_haiku_55_v1: 0.662triage_local_dgx_gemma4_12b_noreason_v1: 0.662triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1: 0.662triage_local_mac_nemotron_35_bolt_e4b_v2: 0.662triage_local_mac_qwen3_30b_a3b_v1: 0.662triage_local_x3_gpt_oss_20b_v1: 0.662triage_haiku_55_v1 · rescored 3d3a6afd: 0.662triage_local_dgx_gemma4_12b_noreason_v1 · rescored 3d3a6afd: 0.662triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 3d3a6afd: 0.662triage_local_mac_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd: 0.662triage_local_mac_qwen3_30b_a3b_v1 · rescored 3d3a6afd: 0.662triage_haiku_55_v1 · rescored 5f2c057c: 0.662triage_local_dgx_gemma4_12b_noreason_v1 · rescored 5f2c057c: 0.662triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 5f2c057c: 0.662triage_local_mac_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c: 0.662triage_local_mac_qwen3_30b_a3b_v1 · rescored 5f2c057c: 0.662triage_haiku_55_v1 · rescored 77adc97b: 0.662triage_local_dgx_gemma4_12b_noreason_v1 · rescored 77adc97b: 0.662triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 77adc97b: 0.662triage_local_mac_qwen3_30b_a3b_v1 · rescored 77adc97b: 0.662triage_haiku_55_v1 · rescored cd35dc1f: 0.662triage_local_dgx_gemma4_12b_noreason_v1 · rescored cd35dc1f: 0.662triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored cd35dc1f: 0.662triage_local_mac_qwen3_30b_a3b_v1 · rescored cd35dc1f: 0.662triage_local_x3_gpt_oss_20b_v1 · rescored cd35dc1f: 0.662triage_hosted_qwen3_30b_a3b_v1: 0.646triage_local_dgx_gpt_oss_120b_v2: 0.646triage_local_dgx_gpt_oss_20b_v1: 0.646triage_local_dgx_qwen3_30b_a3b_v1: 0.646triage_local_x3_nemotron_35_bolt_e4b_v1: 0.646triage_local_x3_nemotron_35_lightning_30b_v1: 0.646triage_local_x3_nemotron_35_bolt_e4b_v1 · repeats: 0.646triage_hosted_qwen3_30b_a3b_v1 · rescored 3d3a6afd: 0.646triage_local_dgx_gpt_oss_120b_v2 · rescored 3d3a6afd: 0.646triage_local_dgx_qwen3_30b_a3b_v1 · rescored 3d3a6afd: 0.646triage_hosted_qwen3_30b_a3b_v1 · rescored 5f2c057c: 0.646triage_local_dgx_gpt_oss_120b_v2 · rescored 5f2c057c: 0.646triage_local_dgx_qwen3_30b_a3b_v1 · rescored 5f2c057c: 0.646triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c: 0.646triage_local_x3_nemotron_35_lightning_30b_v1 · rescored 5f2c057c: 0.646triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c, repeats: 0.646triage_hosted_qwen3_30b_a3b_v1 · rescored 77adc97b: 0.646triage_local_dgx_qwen3_30b_a3b_v1 · rescored 77adc97b: 0.646triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 77adc97b: 0.646triage_local_x3_nemotron_35_lightning_30b_v1 · rescored 77adc97b: 0.646triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 77adc97b, repeats: 0.646triage_hosted_qwen3_30b_a3b_v1 · rescored cd35dc1f: 0.646triage_local_dgx_gpt_oss_20b_v1 · rescored cd35dc1f: 0.646triage_local_dgx_qwen3_30b_a3b_v1 · rescored cd35dc1f: 0.646triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f: 0.646triage_local_x3_nemotron_35_lightning_30b_v1 · rescored cd35dc1f: 0.646triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f, repeats: 0.646triage_gpt_6_luna_v2: 0.631triage_hosted_gpt_oss_120b_v1: 0.631triage_local_mac_gemma4_12b_noreason_v1: 0.631triage_local_mac_gemma4_31b_v1: 0.631triage_haiku_55_v1 · repeats ×3: 0.631triage_gpt_6_luna_v2 · rescored 3d3a6afd: 0.631triage_local_mac_gemma4_12b_noreason_v1 · rescored 3d3a6afd: 0.631triage_haiku_55_v1 · rescored 3d3a6afd, repeats ×3: 0.631triage_gpt_6_luna_v2 · rescored 5f2c057c: 0.631triage_hosted_gpt_oss_120b_v1 · rescored 5f2c057c: 0.631triage_local_mac_gemma4_12b_noreason_v1 · rescored 5f2c057c: 0.631triage_local_mac_gemma4_31b_v1 · rescored 5f2c057c: 0.631triage_haiku_55_v1 · rescored 5f2c057c, repeats ×3: 0.631triage_hosted_gpt_oss_120b_v1 · rescored 77adc97b: 0.631triage_local_mac_gemma4_12b_noreason_v1 · rescored 77adc97b: 0.631triage_local_mac_gemma4_31b_v1 · rescored 77adc97b: 0.631triage_haiku_55_v1 · rescored 77adc97b, repeats ×3: 0.631triage_hosted_gpt_oss_120b_v1 · rescored cd35dc1f: 0.631triage_local_mac_gemma4_12b_noreason_v1 · rescored cd35dc1f: 0.631triage_local_mac_gemma4_31b_v1 · rescored cd35dc1f: 0.631triage_haiku_55_v1 · rescored cd35dc1f, repeats ×3: 0.631triage_gpt_6_luna_v1 · repeats ×3: 0.621triage_gpt_6_luna_v1 · rescored 5f2c057c, repeats ×3: 0.621triage_gpt_6_luna_v1 · rescored 77adc97b, repeats ×3: 0.621triage_gpt_6_luna_v1 · rescored cd35dc1f, repeats ×3: 0.621triage_gemini_35_flash_lite_v1: 0.615triage_hosted_mistral_medium_35_v1: 0.615triage_local_dgx_gemma4_31b_v1: 0.615triage_local_dgx_gpt_oss_120b_v1: 0.615triage_local_mac_gpt_oss_20b_v1: 0.615triage_gemini_35_flash_lite_v1 · rescored 3d3a6afd: 0.615triage_hosted_mistral_medium_35_v1 · rescored 3d3a6afd: 0.615triage_gemini_35_flash_lite_v1 · rescored 5f2c057c: 0.615triage_hosted_mistral_medium_35_v1 · rescored 5f2c057c: 0.615triage_local_dgx_gemma4_31b_v1 · rescored 5f2c057c: 0.615triage_local_dgx_gpt_oss_120b_v1 · rescored 5f2c057c: 0.615triage_gemini_35_flash_lite_v1 · rescored 77adc97b: 0.615triage_hosted_mistral_medium_35_v1 · rescored 77adc97b: 0.615triage_local_dgx_gemma4_31b_v1 · rescored 77adc97b: 0.615triage_local_dgx_gpt_oss_120b_v1 · rescored 77adc97b: 0.615triage_gemini_35_flash_lite_v1 · rescored cd35dc1f: 0.615triage_hosted_mistral_medium_35_v1 · rescored cd35dc1f: 0.615triage_local_dgx_gemma4_31b_v1 · rescored cd35dc1f: 0.615triage_local_dgx_gpt_oss_120b_v1 · rescored cd35dc1f: 0.615triage_local_mac_gpt_oss_20b_v1 · rescored cd35dc1f: 0.615triage_gpt_6_luna_v1: 0.6triage_hosted_gpt_oss_120b_v2: 0.6triage_kimi_k31_pro_v1: 0.6triage_local_dgx_nemotron_35_bolt_e4b_v1: 0.6triage_local_dgx_nemotron_35_lightning_30b_v1: 0.6triage_local_mac_nemotron_35_lightning_30b_v1: 0.6triage_local_x3_gemma4_31b_v1: 0.6triage_gpt_61_sol_v1 · repeats ×3: 0.6triage_hosted_gpt_oss_120b_v2 · rescored 3d3a6afd: 0.6triage_gpt_61_sol_v1 · rescored 3d3a6afd, repeats ×3: 0.6triage_gpt_6_luna_v1 · rescored 5f2c057c: 0.6triage_hosted_gpt_oss_120b_v2 · rescored 5f2c057c: 0.6triage_kimi_k31_pro_v1 · rescored 5f2c057c: 0.6triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c: 0.6triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored 5f2c057c: 0.6triage_local_mac_nemotron_35_lightning_30b_v1 · rescored 5f2c057c: 0.6triage_local_x3_gemma4_31b_v1 · rescored 5f2c057c: 0.6triage_gpt_61_sol_v1 · rescored 5f2c057c, repeats ×3: 0.6triage_gpt_6_luna_v1 · rescored 77adc97b: 0.6triage_kimi_k31_pro_v1 · rescored 77adc97b: 0.6triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored 77adc97b: 0.6triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored 77adc97b: 0.6triage_local_mac_nemotron_35_lightning_30b_v1 · rescored 77adc97b: 0.6triage_local_x3_gemma4_31b_v1 · rescored 77adc97b: 0.6triage_gpt_61_sol_v1 · rescored 77adc97b, repeats ×3: 0.6triage_gpt_6_luna_v1 · rescored cd35dc1f: 0.6triage_kimi_k31_pro_v1 · rescored cd35dc1f: 0.6triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f: 0.6triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored cd35dc1f: 0.6triage_local_mac_nemotron_35_lightning_30b_v1 · rescored cd35dc1f: 0.6triage_local_x3_gemma4_31b_v1 · rescored cd35dc1f: 0.6triage_gpt_61_sol_v1 · rescored cd35dc1f, repeats ×3: 0.6triage_baseline_pro_noreason_v1: 0.585triage_gemini_4_argon_v1: 0.585triage_gpt_61_sol_v1: 0.585triage_kimi_k31_pro_v2: 0.585triage_local_dgx_llama31_8b_v1: 0.585triage_local_dgx_nemotron_35_bolt_e4b_v2: 0.585triage_local_mac_llama31_8b_v1: 0.585triage_local_mac_nemotron_35_bolt_e4b_v1: 0.585triage_local_x3_llama31_8b_v1: 0.585triage_baseline_pro_v1 · repeats ×3: 0.585triage_baseline_pro_noreason_v1 · rescored 3d3a6afd: 0.585triage_gemini_4_argon_v1 · rescored 3d3a6afd: 0.585triage_gpt_61_sol_v1 · rescored 3d3a6afd: 0.585triage_kimi_k31_pro_v2 · rescored 3d3a6afd: 0.585triage_local_dgx_llama31_8b_v1 · rescored 3d3a6afd: 0.585triage_local_dgx_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd: 0.585triage_local_mac_llama31_8b_v1 · rescored 3d3a6afd: 0.585triage_local_x3_llama31_8b_v1 · rescored 3d3a6afd: 0.585triage_baseline_pro_v1 · rescored 3d3a6afd, repeats ×3: 0.585triage_baseline_pro_noreason_v1 · rescored 5f2c057c: 0.585triage_gemini_4_argon_v1 · rescored 5f2c057c: 0.585triage_gpt_61_sol_v1 · rescored 5f2c057c: 0.585triage_kimi_k31_pro_v2 · rescored 5f2c057c: 0.585triage_local_dgx_llama31_8b_v1 · rescored 5f2c057c: 0.585triage_local_dgx_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c: 0.585triage_local_mac_llama31_8b_v1 · rescored 5f2c057c: 0.585triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c: 0.585triage_local_x3_llama31_8b_v1 · rescored 5f2c057c: 0.585triage_baseline_pro_v1 · rescored 5f2c057c, repeats ×3: 0.585triage_gemini_4_argon_v1 · rescored 77adc97b: 0.585triage_gpt_61_sol_v1 · rescored 77adc97b: 0.585triage_local_dgx_llama31_8b_v1 · rescored 77adc97b: 0.585triage_local_mac_llama31_8b_v1 · rescored 77adc97b: 0.585triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored 77adc97b: 0.585triage_local_x3_llama31_8b_v1 · rescored 77adc97b: 0.585triage_baseline_pro_v1 · rescored 77adc97b, repeats ×3: 0.585triage_gemini_4_argon_v1 · rescored cd35dc1f: 0.585triage_gpt_61_sol_v1 · rescored cd35dc1f: 0.585triage_local_dgx_llama31_8b_v1 · rescored cd35dc1f: 0.585triage_local_mac_llama31_8b_v1 · rescored cd35dc1f: 0.585triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f: 0.585triage_local_x3_llama31_8b_v1 · rescored cd35dc1f: 0.585triage_baseline_pro_v1 · rescored cd35dc1f, repeats ×3: 0.585triage_baseline_pro_v1: 0.569triage_hosted_gemma4_31b_v1: 0.569triage_hosted_qwen35_122b_v2: 0.569triage_local_dgx_ministral_3_14b_v1: 0.569triage_local_mac_gpt_oss_20b_v2: 0.569triage_baseline_pro_v1 · rescored 3d3a6afd: 0.569triage_hosted_qwen35_122b_v2 · rescored 3d3a6afd: 0.569triage_local_dgx_ministral_3_14b_v1 · rescored 3d3a6afd: 0.569triage_local_mac_gpt_oss_20b_v2 · rescored 3d3a6afd: 0.569triage_baseline_pro_v1 · rescored 5f2c057c: 0.569triage_hosted_gemma4_31b_v1 · rescored 5f2c057c: 0.569triage_hosted_qwen35_122b_v2 · rescored 5f2c057c: 0.569triage_local_dgx_ministral_3_14b_v1 · rescored 5f2c057c: 0.569triage_local_mac_gpt_oss_20b_v2 · rescored 5f2c057c: 0.569triage_baseline_pro_v1 · rescored 77adc97b: 0.569triage_hosted_gemma4_31b_v1 · rescored 77adc97b: 0.569triage_local_dgx_ministral_3_14b_v1 · rescored 77adc97b: 0.569triage_baseline_pro_v1 · rescored cd35dc1f: 0.569triage_hosted_gemma4_31b_v1 · rescored cd35dc1f: 0.569triage_local_dgx_ministral_3_14b_v1 · rescored cd35dc1f: 0.569triage_local_dgx_gemma4_31b_v2: 0.554triage_local_dgx_mistral_medium_35_128b_v1: 0.554triage_local_x3_gemma4_31b_v2: 0.554triage_local_dgx_gemma4_31b_v2 · rescored 3d3a6afd: 0.554triage_local_dgx_mistral_medium_35_128b_v1 · rescored 3d3a6afd: 0.554triage_local_x3_gemma4_31b_v2 · rescored 3d3a6afd: 0.554triage_local_dgx_gemma4_31b_v2 · rescored 5f2c057c: 0.554triage_local_dgx_mistral_medium_35_128b_v1 · rescored 5f2c057c: 0.554triage_local_x3_gemma4_31b_v2 · rescored 5f2c057c: 0.554triage_local_dgx_mistral_medium_35_128b_v1 · rescored 77adc97b: 0.554triage_local_dgx_mistral_medium_35_128b_v1 · rescored cd35dc1f: 0.554triage_hosted_gemma4_31b_v2: 0.538triage_hosted_qwen35_122b_v1: 0.538triage_local_mac_gemma4_31b_v2: 0.538triage_sonnet_55_v1: 0.538triage_hosted_gemma4_31b_v2 · rescored 3d3a6afd: 0.538triage_local_mac_gemma4_31b_v2 · rescored 3d3a6afd: 0.538triage_sonnet_55_v1 · rescored 3d3a6afd: 0.538triage_hosted_gemma4_31b_v2 · rescored 5f2c057c: 0.538triage_hosted_qwen35_122b_v1 · rescored 5f2c057c: 0.538triage_local_dgx_gpt_oss_20b_v1 · rescored 5f2c057c: 0.538triage_local_mac_gemma4_31b_v2 · rescored 5f2c057c: 0.538triage_sonnet_55_v1 · rescored 5f2c057c: 0.538triage_hosted_qwen35_122b_v1 · rescored 77adc97b: 0.538triage_local_dgx_gpt_oss_20b_v1 · rescored 77adc97b: 0.538triage_sonnet_55_v1 · rescored 77adc97b: 0.538triage_hosted_qwen35_122b_v1 · rescored cd35dc1f: 0.538triage_sonnet_55_v1 · rescored cd35dc1f: 0.538triage_deepseek_41flash_v1: 0.523triage_deepseek_41flash_v2: 0.523triage_local_dgx_gpt_oss_20b_v2: 0.523triage_local_dgx_qwen35_122b_v1: 0.523triage_mistral_large_4_v1: 0.523triage_deepseek_41flash_v2 · rescored 3d3a6afd: 0.523triage_local_dgx_gpt_oss_20b_v2 · rescored 3d3a6afd: 0.523triage_deepseek_41flash_v1 · rescored 5f2c057c: 0.523triage_deepseek_41flash_v2 · rescored 5f2c057c: 0.523triage_local_dgx_gpt_oss_20b_v2 · rescored 5f2c057c: 0.523triage_local_dgx_qwen35_122b_v1 · rescored 5f2c057c: 0.523triage_local_mac_gpt_oss_20b_v1 · rescored 5f2c057c: 0.523triage_local_x3_gpt_oss_20b_v1 · rescored 5f2c057c: 0.523triage_mistral_large_4_v1 · rescored 5f2c057c: 0.523triage_deepseek_41flash_v1 · rescored 77adc97b: 0.523triage_local_dgx_qwen35_122b_v1 · rescored 77adc97b: 0.523triage_local_mac_gpt_oss_20b_v1 · rescored 77adc97b: 0.523triage_local_x3_gpt_oss_20b_v1 · rescored 77adc97b: 0.523triage_mistral_large_4_v1 · rescored 77adc97b: 0.523triage_deepseek_41flash_v1 · rescored cd35dc1f: 0.523triage_local_dgx_qwen35_122b_v1 · rescored cd35dc1f: 0.523triage_mistral_large_4_v1 · rescored cd35dc1f: 0.523triage_local_dgx_phi4_14b_v1: 0.508triage_local_x3_gpt_oss_20b_v2: 0.508triage_mistral_large_4_v2: 0.508triage_local_dgx_phi4_14b_v1 · rescored 3d3a6afd: 0.508triage_local_x3_gpt_oss_20b_v2 · rescored 3d3a6afd: 0.508triage_mistral_large_4_v2 · rescored 3d3a6afd: 0.508triage_local_dgx_phi4_14b_v1 · rescored 5f2c057c: 0.508triage_local_x3_gpt_oss_20b_v2 · rescored 5f2c057c: 0.508triage_mistral_large_4_v2 · rescored 5f2c057c: 0.508triage_local_dgx_phi4_14b_v1 · rescored 77adc97b: 0.508triage_local_dgx_phi4_14b_v1 · rescored cd35dc1f: 0.508triage_local_dgx_qwen38_27b_v1: 0.492triage_local_x3_qwen38_27b_v1: 0.492triage_local_dgx_qwen38_27b_v1 · rescored 5f2c057c: 0.492triage_local_x3_qwen38_27b_v1 · rescored 5f2c057c: 0.492triage_local_dgx_qwen38_27b_v1 · rescored 77adc97b: 0.492triage_local_x3_qwen38_27b_v1 · rescored 77adc97b: 0.492triage_local_dgx_qwen38_27b_v1 · rescored cd35dc1f: 0.492triage_local_x3_qwen38_27b_v1 · rescored cd35dc1f: 0.492triage_local_dgx_nemotron_35_super_120b_v2: 0.477triage_local_mac_ministral_3_14b_v1: 0.477triage_local_mac_nemotron_35_lightning_30b_v2: 0.477triage_local_mac_phi4_14b_v1: 0.477triage_local_mac_qwen38_27b_v1: 0.477triage_local_x3_ministral_3_14b_v1: 0.477triage_local_x3_phi4_14b_v1: 0.477triage_local_dgx_nemotron_35_super_120b_v2 · rescored 3d3a6afd: 0.477triage_local_mac_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd: 0.477triage_local_mac_phi4_14b_v1 · rescored 3d3a6afd: 0.477triage_local_x3_ministral_3_14b_v1 · rescored 3d3a6afd: 0.477triage_local_x3_phi4_14b_v1 · rescored 3d3a6afd: 0.477triage_local_dgx_nemotron_35_super_120b_v2 · rescored 5f2c057c: 0.477triage_local_mac_nemotron_35_lightning_30b_v2 · rescored 5f2c057c: 0.477triage_local_mac_phi4_14b_v1 · rescored 5f2c057c: 0.477triage_local_mac_qwen38_27b_v1 · rescored 5f2c057c: 0.477triage_local_x3_ministral_3_14b_v1 · rescored 5f2c057c: 0.477triage_local_x3_phi4_14b_v1 · rescored 5f2c057c: 0.477triage_local_mac_phi4_14b_v1 · rescored 77adc97b: 0.477triage_local_mac_qwen38_27b_v1 · rescored 77adc97b: 0.477triage_local_x3_ministral_3_14b_v1 · rescored 77adc97b: 0.477triage_local_x3_phi4_14b_v1 · rescored 77adc97b: 0.477triage_local_mac_ministral_3_14b_v1 · rescored cd35dc1f: 0.477triage_local_mac_phi4_14b_v1 · rescored cd35dc1f: 0.477triage_local_mac_qwen38_27b_v1 · rescored cd35dc1f: 0.477triage_local_x3_ministral_3_14b_v1 · rescored cd35dc1f: 0.477triage_local_x3_phi4_14b_v1 · rescored cd35dc1f: 0.477triage_local_dgx_nemotron_35_super_120b_v1: 0.462triage_local_mac_ministral_3_14b_v1 · rescored 3d3a6afd: 0.462triage_local_dgx_nemotron_35_super_120b_v1 · rescored 5f2c057c: 0.462triage_local_mac_ministral_3_14b_v1 · rescored 5f2c057c: 0.462triage_local_dgx_nemotron_35_super_120b_v1 · rescored 77adc97b: 0.462triage_local_mac_ministral_3_14b_v1 · rescored 77adc97b: 0.462triage_local_dgx_nemotron_35_super_120b_v1 · rescored cd35dc1f: 0.462triage_local_dgx_nemotron_35_lightning_30b_v2: 0.446triage_local_dgx_qwen38_27b_v2: 0.446triage_local_mac_qwen38_27b_v2: 0.446triage_local_x3_nemotron_35_lightning_30b_v2: 0.446triage_local_dgx_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd: 0.446triage_local_dgx_qwen38_27b_v2 · rescored 3d3a6afd: 0.446triage_local_mac_qwen38_27b_v2 · rescored 3d3a6afd: 0.446triage_local_x3_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd: 0.446triage_local_dgx_nemotron_35_lightning_30b_v2 · rescored 5f2c057c: 0.446triage_local_dgx_qwen38_27b_v2 · rescored 5f2c057c: 0.446triage_local_mac_qwen38_27b_v2 · rescored 5f2c057c: 0.446triage_local_x3_nemotron_35_lightning_30b_v2 · rescored 5f2c057c: 0.446triage_local_x3_qwen38_27b_v2: 0.415triage_local_x3_qwen38_27b_v2 · rescored 3d3a6afd: 0.415triage_local_x3_qwen38_27b_v2 · rescored 5f2c057c: 0.415triage_local_dgx_llama4_scout_v1: 0.385triage_local_dgx_qwen35_122b_v2: 0.385triage_local_dgx_llama4_scout_v1 · rescored 3d3a6afd: 0.385triage_local_dgx_qwen35_122b_v2 · rescored 3d3a6afd: 0.385triage_local_dgx_llama4_scout_v1 · rescored 5f2c057c: 0.385triage_local_dgx_qwen35_122b_v2 · rescored 5f2c057c: 0.385triage_local_dgx_llama4_scout_v1 · rescored 77adc97b: 0.385triage_local_dgx_llama4_scout_v1 · rescored cd35dc1f: 0.385triage_stub_dismiss_v1: 0.369triage_stub_dismiss_v1 · rescored 3d3a6afd: 0.369triage_stub_dismiss_v1 · rescored 5f2c057c: 0.369triage_stub_dismiss_v1 · rescored 77adc97b: 0.369triage_stub_dismiss_v1 · rescored cd35dc1f: 0.369triage_stub_escalate_v1: 0.354triage_stub_escalate_v1 · rescored 3d3a6afd: 0.354triage_stub_escalate_v1 · rescored 5f2c057c: 0.354triage_stub_escalate_v1 · rescored 77adc97b: 0.354triage_stub_escalate_v1 · rescored cd35dc1f: 0.3540.354better · 0.692

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 correctitemsscored against
triage_local_x3_qwen3_30b_a3b_v10.69265gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen3_30b_a3b_v1 · rescored 3d3a6afd0.69265gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen3_30b_a3b_v1 · rescored 5f2c057c0.69265gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen3_30b_a3b_v1 · rescored 77adc97b0.69265gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen3_30b_a3b_v1 · rescored cd35dc1f0.69265gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v20.67765gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd0.67765gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c0.67765gold/orrery_triage_v1_labelled gold
triage_haiku_55_v10.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v10.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v10.66265gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v20.66265gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v10.66265gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v10.66265gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 3d3a6afd0.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v1 · rescored 3d3a6afd0.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 3d3a6afd0.66265gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd0.66265gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v1 · rescored 3d3a6afd0.66265gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 5f2c057c0.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v1 · rescored 5f2c057c0.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 5f2c057c0.66265gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c0.66265gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v1 · rescored 5f2c057c0.66265gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 77adc97b0.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v1 · rescored 77adc97b0.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored 77adc97b0.66265gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v1 · rescored 77adc97b0.66265gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored cd35dc1f0.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_12b_noreason_v1 · rescored cd35dc1f0.66265gold/orrery_triage_v1_labelled gold
triage_local_dgx_vllm_qwen3_30b_a3b_nvfp4_v1 · rescored cd35dc1f0.66265gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen3_30b_a3b_v1 · rescored cd35dc1f0.66265gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v1 · rescored cd35dc1f0.66265gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v10.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v20.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v10.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v10.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v10.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v10.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · repeats0.64665gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v1 · rescored 3d3a6afd0.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v2 · rescored 3d3a6afd0.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v1 · rescored 3d3a6afd0.64665gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v1 · rescored 5f2c057c0.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v2 · rescored 5f2c057c0.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v1 · rescored 5f2c057c0.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c0.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v1 · rescored 5f2c057c0.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c, repeats0.64665gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v1 · rescored 77adc97b0.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v1 · rescored 77adc97b0.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 77adc97b0.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v1 · rescored 77adc97b0.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored 77adc97b, repeats0.64665gold/orrery_triage_v1_labelled gold
triage_hosted_qwen3_30b_a3b_v1 · rescored cd35dc1f0.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v1 · rescored cd35dc1f0.64665gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen3_30b_a3b_v1 · rescored cd35dc1f0.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f0.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v1 · rescored cd35dc1f0.64665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f, repeats0.64665gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v20.63165gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v10.63165gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v10.63165gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v10.63165gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · repeats ×30.63165gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v2 · rescored 3d3a6afd0.63165gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v1 · rescored 3d3a6afd0.63165gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 3d3a6afd, repeats ×30.63165gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v2 · rescored 5f2c057c0.63165gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v1 · rescored 5f2c057c0.63165gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v1 · rescored 5f2c057c0.63165gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v1 · rescored 5f2c057c0.63165gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 5f2c057c, repeats ×30.63165gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v1 · rescored 77adc97b0.63165gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v1 · rescored 77adc97b0.63165gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v1 · rescored 77adc97b0.63165gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored 77adc97b, repeats ×30.63165gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v1 · rescored cd35dc1f0.63165gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_12b_noreason_v1 · rescored cd35dc1f0.63165gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v1 · rescored cd35dc1f0.63165gold/orrery_triage_v1_labelled gold
triage_haiku_55_v1 · rescored cd35dc1f, repeats ×30.63165gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · repeats ×30.62165gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored 5f2c057c, repeats ×30.62165gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored 77adc97b, repeats ×30.62165gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored cd35dc1f, repeats ×30.62165gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v10.61565gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v10.61565gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v10.61565gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v10.61565gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v10.61565gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v1 · rescored 3d3a6afd0.61565gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v1 · rescored 3d3a6afd0.61565gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v1 · rescored 5f2c057c0.61565gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v1 · rescored 5f2c057c0.61565gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v1 · rescored 5f2c057c0.61565gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v1 · rescored 5f2c057c0.61565gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v1 · rescored 77adc97b0.61565gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v1 · rescored 77adc97b0.61565gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v1 · rescored 77adc97b0.61565gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v1 · rescored 77adc97b0.61565gold/orrery_triage_v1_labelled gold
triage_gemini_35_flash_lite_v1 · rescored cd35dc1f0.61565gold/orrery_triage_v1_labelled gold
triage_hosted_mistral_medium_35_v1 · rescored cd35dc1f0.61565gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v1 · rescored cd35dc1f0.61565gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_120b_v1 · rescored cd35dc1f0.61565gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v1 · rescored cd35dc1f0.61565gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v10.60065gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v20.60065gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v10.60065gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v10.60065gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v10.60065gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v10.60065gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v10.60065gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · repeats ×30.60065gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v2 · rescored 3d3a6afd0.60065gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 3d3a6afd, repeats ×30.60065gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored 5f2c057c0.60065gold/orrery_triage_v1_labelled gold
triage_hosted_gpt_oss_120b_v2 · rescored 5f2c057c0.60065gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v1 · rescored 5f2c057c0.60065gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c0.60065gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored 5f2c057c0.60065gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v1 · rescored 5f2c057c0.60065gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v1 · rescored 5f2c057c0.60065gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 5f2c057c, repeats ×30.60065gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored 77adc97b0.60065gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v1 · rescored 77adc97b0.60065gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored 77adc97b0.60065gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored 77adc97b0.60065gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v1 · rescored 77adc97b0.60065gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v1 · rescored 77adc97b0.60065gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 77adc97b, repeats ×30.60065gold/orrery_triage_v1_labelled gold
triage_gpt_6_luna_v1 · rescored cd35dc1f0.60065gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v1 · rescored cd35dc1f0.60065gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f0.60065gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v1 · rescored cd35dc1f0.60065gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v1 · rescored cd35dc1f0.60065gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v1 · rescored cd35dc1f0.60065gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored cd35dc1f, repeats ×30.60065gold/orrery_triage_v1_labelled gold
triage_baseline_pro_noreason_v10.58565gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v10.58565gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v10.58565gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v20.58565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v10.58565gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v20.58565gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v10.58565gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v10.58565gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v10.58565gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · repeats ×30.58565gold/orrery_triage_v1_labelled gold
triage_baseline_pro_noreason_v1 · rescored 3d3a6afd0.58565gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v1 · rescored 3d3a6afd0.58565gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 3d3a6afd0.58565gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v2 · rescored 3d3a6afd0.58565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v1 · rescored 3d3a6afd0.58565gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v2 · rescored 3d3a6afd0.58565gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v1 · rescored 3d3a6afd0.58565gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v1 · rescored 3d3a6afd0.58565gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 3d3a6afd, repeats ×30.58565gold/orrery_triage_v1_labelled gold
triage_baseline_pro_noreason_v1 · rescored 5f2c057c0.58565gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v1 · rescored 5f2c057c0.58565gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 5f2c057c0.58565gold/orrery_triage_v1_labelled gold
triage_kimi_k31_pro_v2 · rescored 5f2c057c0.58565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v1 · rescored 5f2c057c0.58565gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_bolt_e4b_v2 · rescored 5f2c057c0.58565gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v1 · rescored 5f2c057c0.58565gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored 5f2c057c0.58565gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v1 · rescored 5f2c057c0.58565gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 5f2c057c, repeats ×30.58565gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v1 · rescored 77adc97b0.58565gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored 77adc97b0.58565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v1 · rescored 77adc97b0.58565gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v1 · rescored 77adc97b0.58565gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored 77adc97b0.58565gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v1 · rescored 77adc97b0.58565gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 77adc97b, repeats ×30.58565gold/orrery_triage_v1_labelled gold
triage_gemini_4_argon_v1 · rescored cd35dc1f0.58565gold/orrery_triage_v1_labelled gold
triage_gpt_61_sol_v1 · rescored cd35dc1f0.58565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama31_8b_v1 · rescored cd35dc1f0.58565gold/orrery_triage_v1_labelled gold
triage_local_mac_llama31_8b_v1 · rescored cd35dc1f0.58565gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_bolt_e4b_v1 · rescored cd35dc1f0.58565gold/orrery_triage_v1_labelled gold
triage_local_x3_llama31_8b_v1 · rescored cd35dc1f0.58565gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored cd35dc1f, repeats ×30.58565gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v10.56965gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v10.56965gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v20.56965gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v10.56965gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v20.56965gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 3d3a6afd0.56965gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v2 · rescored 3d3a6afd0.56965gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v1 · rescored 3d3a6afd0.56965gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v2 · rescored 3d3a6afd0.56965gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 5f2c057c0.56965gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v1 · rescored 5f2c057c0.56965gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v2 · rescored 5f2c057c0.56965gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v1 · rescored 5f2c057c0.56965gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v2 · rescored 5f2c057c0.56965gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored 77adc97b0.56965gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v1 · rescored 77adc97b0.56965gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v1 · rescored 77adc97b0.56965gold/orrery_triage_v1_labelled gold
triage_baseline_pro_v1 · rescored cd35dc1f0.56965gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v1 · rescored cd35dc1f0.56965gold/orrery_triage_v1_labelled gold
triage_local_dgx_ministral_3_14b_v1 · rescored cd35dc1f0.56965gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v20.55465gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v10.55465gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v20.55465gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v2 · rescored 3d3a6afd0.55465gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v1 · rescored 3d3a6afd0.55465gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v2 · rescored 3d3a6afd0.55465gold/orrery_triage_v1_labelled gold
triage_local_dgx_gemma4_31b_v2 · rescored 5f2c057c0.55465gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v1 · rescored 5f2c057c0.55465gold/orrery_triage_v1_labelled gold
triage_local_x3_gemma4_31b_v2 · rescored 5f2c057c0.55465gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v1 · rescored 77adc97b0.55465gold/orrery_triage_v1_labelled gold
triage_local_dgx_mistral_medium_35_128b_v1 · rescored cd35dc1f0.55465gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v20.53865gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v10.53865gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v20.53865gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v10.53865gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v2 · rescored 3d3a6afd0.53865gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v2 · rescored 3d3a6afd0.53865gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v1 · rescored 3d3a6afd0.53865gold/orrery_triage_v1_labelled gold
triage_hosted_gemma4_31b_v2 · rescored 5f2c057c0.53865gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v1 · rescored 5f2c057c0.53865gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v1 · rescored 5f2c057c0.53865gold/orrery_triage_v1_labelled gold
triage_local_mac_gemma4_31b_v2 · rescored 5f2c057c0.53865gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v1 · rescored 5f2c057c0.53865gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v1 · rescored 77adc97b0.53865gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v1 · rescored 77adc97b0.53865gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v1 · rescored 77adc97b0.53865gold/orrery_triage_v1_labelled gold
triage_hosted_qwen35_122b_v1 · rescored cd35dc1f0.53865gold/orrery_triage_v1_labelled gold
triage_sonnet_55_v1 · rescored cd35dc1f0.53865gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v10.52365gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v20.52365gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v20.52365gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v10.52365gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v10.52365gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v2 · rescored 3d3a6afd0.52365gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v2 · rescored 3d3a6afd0.52365gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v1 · rescored 5f2c057c0.52365gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v2 · rescored 5f2c057c0.52365gold/orrery_triage_v1_labelled gold
triage_local_dgx_gpt_oss_20b_v2 · rescored 5f2c057c0.52365gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v1 · rescored 5f2c057c0.52365gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v1 · rescored 5f2c057c0.52365gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v1 · rescored 5f2c057c0.52365gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v1 · rescored 5f2c057c0.52365gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v1 · rescored 77adc97b0.52365gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v1 · rescored 77adc97b0.52365gold/orrery_triage_v1_labelled gold
triage_local_mac_gpt_oss_20b_v1 · rescored 77adc97b0.52365gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v1 · rescored 77adc97b0.52365gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v1 · rescored 77adc97b0.52365gold/orrery_triage_v1_labelled gold
triage_deepseek_41flash_v1 · rescored cd35dc1f0.52365gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v1 · rescored cd35dc1f0.52365gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v1 · rescored cd35dc1f0.52365gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v10.50865gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v20.50865gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v20.50865gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v1 · rescored 3d3a6afd0.50865gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v2 · rescored 3d3a6afd0.50865gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v2 · rescored 3d3a6afd0.50865gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v1 · rescored 5f2c057c0.50865gold/orrery_triage_v1_labelled gold
triage_local_x3_gpt_oss_20b_v2 · rescored 5f2c057c0.50865gold/orrery_triage_v1_labelled gold
triage_mistral_large_4_v2 · rescored 5f2c057c0.50865gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v1 · rescored 77adc97b0.50865gold/orrery_triage_v1_labelled gold
triage_local_dgx_phi4_14b_v1 · rescored cd35dc1f0.50865gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v10.49265gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v10.49265gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v1 · rescored 5f2c057c0.49265gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v1 · rescored 5f2c057c0.49265gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v1 · rescored 77adc97b0.49265gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v1 · rescored 77adc97b0.49265gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v1 · rescored cd35dc1f0.49265gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v1 · rescored cd35dc1f0.49265gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v20.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v10.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v20.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v10.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v10.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v10.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v10.47765gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v2 · rescored 3d3a6afd0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v1 · rescored 3d3a6afd0.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v1 · rescored 3d3a6afd0.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v1 · rescored 3d3a6afd0.47765gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v2 · rescored 5f2c057c0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_nemotron_35_lightning_30b_v2 · rescored 5f2c057c0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v1 · rescored 5f2c057c0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v1 · rescored 5f2c057c0.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v1 · rescored 5f2c057c0.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v1 · rescored 5f2c057c0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v1 · rescored 77adc97b0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v1 · rescored 77adc97b0.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v1 · rescored 77adc97b0.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v1 · rescored 77adc97b0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v1 · rescored cd35dc1f0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_phi4_14b_v1 · rescored cd35dc1f0.47765gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v1 · rescored cd35dc1f0.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_ministral_3_14b_v1 · rescored cd35dc1f0.47765gold/orrery_triage_v1_labelled gold
triage_local_x3_phi4_14b_v1 · rescored cd35dc1f0.47765gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v10.46265gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v1 · rescored 3d3a6afd0.46265gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v1 · rescored 5f2c057c0.46265gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v1 · rescored 5f2c057c0.46265gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v1 · rescored 77adc97b0.46265gold/orrery_triage_v1_labelled gold
triage_local_mac_ministral_3_14b_v1 · rescored 77adc97b0.46265gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_super_120b_v1 · rescored cd35dc1f0.46265gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v20.44665gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v20.44665gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v20.44665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v20.44665gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd0.44665gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v2 · rescored 3d3a6afd0.44665gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v2 · rescored 3d3a6afd0.44665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v2 · rescored 3d3a6afd0.44665gold/orrery_triage_v1_labelled gold
triage_local_dgx_nemotron_35_lightning_30b_v2 · rescored 5f2c057c0.44665gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen38_27b_v2 · rescored 5f2c057c0.44665gold/orrery_triage_v1_labelled gold
triage_local_mac_qwen38_27b_v2 · rescored 5f2c057c0.44665gold/orrery_triage_v1_labelled gold
triage_local_x3_nemotron_35_lightning_30b_v2 · rescored 5f2c057c0.44665gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v20.41565gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v2 · rescored 3d3a6afd0.41565gold/orrery_triage_v1_labelled gold
triage_local_x3_qwen38_27b_v2 · rescored 5f2c057c0.41565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v10.38565gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v20.38565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v1 · rescored 3d3a6afd0.38565gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v2 · rescored 3d3a6afd0.38565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v1 · rescored 5f2c057c0.38565gold/orrery_triage_v1_labelled gold
triage_local_dgx_qwen35_122b_v2 · rescored 5f2c057c0.38565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v1 · rescored 77adc97b0.38565gold/orrery_triage_v1_labelled gold
triage_local_dgx_llama4_scout_v1 · rescored cd35dc1f0.38565gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v10.36965gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v1 · rescored 3d3a6afd0.36965gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v1 · rescored 5f2c057c0.36965gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v1 · rescored 77adc97b0.36965gold/orrery_triage_v1_labelled gold
triage_stub_dismiss_v1 · rescored cd35dc1f0.36965gold/orrery_triage_v1_labelled gold
triage_stub_escalate_v10.35465gold/orrery_triage_v1_labelled gold
triage_stub_escalate_v1 · rescored 3d3a6afd0.35465gold/orrery_triage_v1_labelled gold
triage_stub_escalate_v1 · rescored 5f2c057c0.35465gold/orrery_triage_v1_labelled gold
triage_stub_escalate_v1 · rescored 77adc97b0.35465gold/orrery_triage_v1_labelled gold
triage_stub_escalate_v1 · rescored cd35dc1f0.35465gold/orrery_triage_v1_labelled gold