{ "context": { "cap_tokens": 261056, "max_model_len": 262144, "targets": [ { "attempted_context_tokens": 32768, "chars_per_token": 3.0, "clamped_tokens": 32768, "e2e_ms": 2954.9683000004734, "estimated_prompt_tokens": 32768, "output_tokens": 50, "status": "passed", "target_tokens": 32768, "ttft_ms": 2266.3140000004205, "usage": { "completion_tokens": 50, "prompt_tokens": 31254, "total_tokens": 31304 } } ] }, "evaluation_protocol": { "budget_semantics": "visible_answer_tokens plus reasoning_headroom_tokens are sent as one max_tokens cap; the endpoint does not hard-partition the channels", "minimum_comparable_repetitions": 3, "minimum_pass_rate": 1.0, "reasoning_headroom_tokens": 0, "records_finish_reason": true, "records_full_visible_answer": true, "records_reasoning_channel_metadata": true, "repetitions": 3, "version": 3, "visible_answer_tokens": 512 }, "failures": [], "identity": { "base_url": "http://127.0.0.1:39054/v1", "candidate_id": "laguna-s-2.1-nvfp4-tp2", "config_id": "tp2-thinking-disabled", "engine": "vllm-0.25.1", "gpu": "dual-rtx-pro-6000-blackwell-max-q-tp2-pcie", "model": "laguna-s-2.1-nvfp4-tp2", "started_at": "2026-08-01T13:08:03Z" }, "intelligence": { "checks": [ { "attempt_count": 3, "attempts": [ { "attempt": 1, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```", "content_excerpt": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```", "failure_class": null, "finish_reason": "stop", "latency_ms": 667.2214000009262, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "text_checks": [ { "name": "diff_shape", "passed": true }, { "name": "removes_old_timeout", "passed": true }, { "name": "adds_new_timeout", "passed": true } ], "usage": { "completion_tokens": 44, "prompt_tokens": 79, "prompt_tokens_details": null, "total_tokens": 123 } }, { "attempt": 2, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```", "content_excerpt": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```", "failure_class": null, "finish_reason": "stop", "latency_ms": 659.7907000013947, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "text_checks": [ { "name": "diff_shape", "passed": true }, { "name": "removes_old_timeout", "passed": true }, { "name": "adds_new_timeout", "passed": true } ], "usage": { "completion_tokens": 44, "prompt_tokens": 79, "prompt_tokens_details": null, "total_tokens": 123 } }, { "attempt": 3, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```", "content_excerpt": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```", "failure_class": null, "finish_reason": "stop", "latency_ms": 650.5893999983527, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "text_checks": [ { "name": "diff_shape", "passed": true }, { "name": "removes_old_timeout", "passed": true }, { "name": "adds_new_timeout", "passed": true } ], "usage": { "completion_tokens": 44, "prompt_tokens": 79, "prompt_tokens_details": null, "total_tokens": 123 } } ], "id": "unified_diff_timeout_edit", "pass_count": 3, "pass_rate": 1.0, "required_pass_rate": 1.0, "status": "passed", "validator": "deterministic_text_checks" }, { "attempt_count": 3, "attempts": [ { "attempt": 1, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "The problem is that the combined STT+LLM+TTS latency (550+1800+650 = 3000 ms) exceeds the 2500 ms turn timeout, causing the agent to cut off before completing; a practical fix is to reduce the LLM latency (e.g., use a faster model or shorter prompt) to bring the total under 2500 ms.", "content_excerpt": "The problem is that the combined STT+LLM+TTS latency (550+1800+650 = 3000 ms) exceeds the 2500 ms turn timeout, causing the agent to cut off before completing; a practical fix is to reduce the LLM lat", "failure_class": null, "finish_reason": "stop", "latency_ms": 1343.7176999996154, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "text_checks": [ { "name": "identifies_budget_overrun", "passed": true }, { "name": "offers_latency_fix", "passed": true } ], "usage": { "completion_tokens": 90, "prompt_tokens": 106, "prompt_tokens_details": null, "total_tokens": 196 } }, { "attempt": 2, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "The problem is that the combined STT (550 ms) + LLM (1800 ms) + TTS (650 ms) = 2950 ms exceeds the 2500 ms turn timeout, causing the agent to time out. A practical fix is to reduce the LLM latency (e.g., by using a faster model or optimizing prompts) to bring the total under 2500 ms.", "content_excerpt": "The problem is that the combined STT (550 ms) + LLM (1800 ms) + TTS (650 ms) = 2950 ms exceeds the 2500 ms turn timeout, causing the agent to time out. A practical fix is to reduce the LLM latency (e.", "failure_class": null, "finish_reason": "stop", "latency_ms": 1360.953100000188, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "text_checks": [ { "name": "identifies_budget_overrun", "passed": true }, { "name": "offers_latency_fix", "passed": true } ], "usage": { "completion_tokens": 93, "prompt_tokens": 106, "prompt_tokens_details": null, "total_tokens": 199 } }, { "attempt": 3, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "The problem is that the combined STT (550 ms) + LLM (1800 ms) + TTS (650 ms) = 3000 ms exceeds the 2500 ms total timeout, causing the agent to fail. A practical fix is to reduce the LLM latency (e.g., by using a faster model or optimizing prompts) to bring the total under 2500 ms.", "content_excerpt": "The problem is that the combined STT (550 ms) + LLM (1800 ms) + TTS (650 ms) = 3000 ms exceeds the 2500 ms total timeout, causing the agent to fail. A practical fix is to reduce the LLM latency (e.g.,", "failure_class": null, "finish_reason": "stop", "latency_ms": 1324.9771999999211, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "text_checks": [ { "name": "identifies_budget_overrun", "passed": true }, { "name": "offers_latency_fix", "passed": true } ], "usage": { "completion_tokens": 92, "prompt_tokens": 106, "prompt_tokens_details": null, "total_tokens": 198 } } ], "id": "parallel_timeout_triage", "pass_count": 3, "pass_rate": 1.0, "required_pass_rate": 1.0, "status": "passed", "validator": "deterministic_text_checks" } ], "status": "passed" }, "run_id": "fast-bakeoff-20260801T130803Z", "schema": "anvil-serving.fast-tier-bakeoff/v1", "score_inputs": { "e2e_p50_ms": 2954.9683000004734, "intelligence_pass_rate": 1.0, "operational_fit_notes": [ "endpoint was already loaded; benchmark did not start or stop serves" ], "session_recall_passed": true, "thinking_mode": "disabled", "tool_call_passed": true, "ttft_p50_ms": 2266.3140000004205, "usable_context_tokens": 32768, "voice_latency_ms": null }, "selection": { "context_targets": [ 32768 ], "endpoint_already_loaded": true, "requests_per_context": 1, "suites": [ "chat", "context", "intelligence", "session", "tool" ] }, "session": { "checks": [ { "attempt_count": 3, "attempts": [ { "attempt": 1, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "RIVER-918", "content_excerpt": "RIVER-918", "expected": "RIVER-918", "failure_class": null, "finish_reason": "stop", "latency_ms": 129.55420000071172, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "usage": { "completion_tokens": 6, "prompt_tokens": 81, "prompt_tokens_details": null, "total_tokens": 87 } }, { "attempt": 2, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "RIVER-918", "content_excerpt": "RIVER-918", "expected": "RIVER-918", "failure_class": null, "finish_reason": "stop", "latency_ms": 101.1636000002909, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "usage": { "completion_tokens": 6, "prompt_tokens": 81, "prompt_tokens_details": null, "total_tokens": 87 } }, { "attempt": 3, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "RIVER-918", "content_excerpt": "RIVER-918", "expected": "RIVER-918", "failure_class": null, "finish_reason": "stop", "latency_ms": 99.70170000087819, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "usage": { "completion_tokens": 6, "prompt_tokens": 81, "prompt_tokens_details": null, "total_tokens": 87 } } ], "name": "single_request_multiturn_recall", "pass_count": 3, "pass_rate": 1.0, "required_pass_rate": 1.0, "status": "passed" } ], "status": "passed" }, "source_recipe": { "ref": "configs/tp2-model-campaign-recipes.toml", "serve_command": null }, "suites": {}, "thinking": { "chat_template_kwargs": { "enable_thinking": false }, "control_evidence": "docs/findings/2026-08-01-dual-pro-tp2-campaign-evidence/laguna-s-21-thinking-disabled-control.json", "control_evidence_sha256": "407399b75de75d68125260ed2d68c58e361c9c5e323a72821721a598f3d5d31d", "control_mechanism": "chat_template_kwargs", "control_requested": { "enable_thinking": false }, "control_status": "verified", "mode": "disabled", "reasoning_effort": null, "unsupported": false }, "timing": { "chat": { "content_chunks": 49, "decode_tok_s_p50": 71.15326223911799, "decode_tok_s_p95": 71.15326223911799, "e2e_p50_ms": 2954.9683000004734, "e2e_p95_ms": 2954.9683000004734, "effective_prefill_tok_s_p50": 13790.675078561135, "effective_prefill_tok_s_p95": 13790.675078561135, "generation_p50_ms": 688.6543000000529, "generation_p95_ms": 688.6543000000529, "mean_inter_token_latency_ms_p50": 14.05416938775618, "mean_inter_token_latency_ms_p95": 14.05416938775618, "output_tokens": 50, "prompt_token_samples": 1, "prompt_tokens": 31254, "prompt_tokens_p50": 31254.0, "prompt_tokens_p95": 31254.0, "ttft_p50_ms": 2266.3140000004205, "ttft_p95_ms": 2266.3140000004205, "ttft_samples": 1 }, "wall_ms": 10610.501199998907 }, "tool": { "checks": [ { "attempt_count": 3, "attempts": [ { "arguments": { "zip": "98101" }, "attempt": 1, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "", "content_excerpt": "", "failure_class": null, "finish_reason": "tool_calls", "latency_ms": 436.04659999982687, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "tool_call_count": 1, "usage": { "completion_tokens": 27, "prompt_tokens": 149, "prompt_tokens_details": null, "total_tokens": 176 }, "valid_tool_call_count": 1, "validation_errors": [] }, { "arguments": { "zip": "98101" }, "attempt": 2, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "", "content_excerpt": "", "failure_class": null, "finish_reason": "tool_calls", "latency_ms": 429.71810000017285, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "tool_call_count": 1, "usage": { "completion_tokens": 27, "prompt_tokens": 149, "prompt_tokens_details": null, "total_tokens": 176 }, "valid_tool_call_count": 1, "validation_errors": [] }, { "arguments": { "zip": "98101" }, "attempt": 3, "budget": { "legacy_max_tokens_as_visible": false, "max_completion_tokens": 512, "reasoning_headroom_tokens": 0, "visible_answer_tokens": 512 }, "content": "", "content_excerpt": "", "failure_class": null, "finish_reason": "tool_calls", "latency_ms": 420.04109999834327, "reasoning_chars": 0, "reasoning_excerpt": "", "reasoning_field": null, "reasoning_tokens": null, "status": "passed", "tool_call_count": 1, "usage": { "completion_tokens": 27, "prompt_tokens": 149, "prompt_tokens_details": null, "total_tokens": 176 }, "valid_tool_call_count": 1, "validation_errors": [] } ], "expected_arguments": { "zip": "98101" }, "expected_function": "record_weather_zip", "name": "openai_tool_call_smoke", "pass_count": 3, "pass_rate": 1.0, "required_pass_rate": 1.0, "status": "passed" } ], "status": "passed" }, "voice": { "llm_latency_ms": null, "status": "not_run", "stt_latency_ms": null, "total_turn_latency_ms": null, "tts_latency_ms": null } }