Nothing deleted, everything hashed
Evidence
Every run — accepted, failed, or invalid — produces a manifest: raw request, raw response, parsed output, usage, cost, validation, trace, and a SHA-256 evidence hash over the artifacts. Failed runs are kept and cited.
30 manifests · real runs only · how hashing works
run-real-flagship-fight-anthropic-claude-opus-4-5-001
anthropic-claude-opus-4-5 · model · workload v1.0.0
SHA-256 bb397bd6a3714801807d0ee7e3ccaad566661cb68fc3763b57f7fdb793e84d00
run-real-flagship-fight-anthropic-claude-opus-4-5-002
anthropic-claude-opus-4-5 · model · workload v1.0.0
SHA-256 e5cc1771226f5e5695461d16a462200bfb65255dc1be353778bf01691955c88f
run-real-flagship-fight-anthropic-claude-opus-4-5-003
anthropic-claude-opus-4-5 · model · workload v1.0.0
SHA-256 bfce4f726579b4bc7d081769c22e99833e2f998e5fe120561a33ad650ba3c018
run-real-flagship-fight-cohere-command-a-001
cohere-command-a · model · workload v1.0.0
SHA-256 b7fdc86ebe9f889ccf1e7e16f9eb0b6140ef41fd61377976e55d688c1967651c
run-real-flagship-fight-cohere-command-a-002
cohere-command-a · model · workload v1.0.0
SHA-256 47764139f1ad0468d8a7e4529316b0dbe10e177c888572efdcb5cd8836ae4671
run-real-flagship-fight-cohere-command-a-003
cohere-command-a · model · workload v1.0.0
SHA-256 e0f6636296adc631f3d99b78413bad2ccc3f9a22be9419e161e53b5f0a2e9ba5
run-real-flagship-fight-deepseek-deepseek-v3-2-001
deepseek-deepseek-v3-2 · model · workload v1.0.0
SHA-256 99335f83ed18ae97a0ce5558916ecf24db30d79cbb6f29c83e013519b60ddb00
run-real-flagship-fight-deepseek-deepseek-v3-2-002
deepseek-deepseek-v3-2 · model · workload v1.0.0
SHA-256 ceb23d790bdc0b1dfaa4e421697a0ac4a8c94fd375a84d4b38ccce131fbbcfa3
run-real-flagship-fight-deepseek-deepseek-v3-2-003
deepseek-deepseek-v3-2 · model · workload v1.0.0
SHA-256 8cd8b1c002433dd662dfe5398eaa0f5277d87a074c633d3dc48c369249d8abde
run-real-flagship-fight-google-gemini-3-1-pro-preview-001
google-gemini-3-1-pro-preview · model · workload v1.0.0
SHA-256 ba7b843dcf5b6de65a97518f0dcbf7f6de7b477081362ea1ed447622cdc86867
run-real-flagship-fight-google-gemini-3-1-pro-preview-002
google-gemini-3-1-pro-preview · model · workload v1.0.0
SHA-256 365944043307928e1a5f223c0665ced6fe42ca63bb8c3cd1f751a409a22aae19
run-real-flagship-fight-google-gemini-3-1-pro-preview-003
google-gemini-3-1-pro-preview · model · workload v1.0.0
SHA-256 6b6e1e567d3bb55c286921335ff77a5fac3a40b06f06ec68ffc544ee59139be9
run-real-flagship-fight-meta-llama-4-maverick-001
meta-llama-4-maverick · model · workload v1.0.0
SHA-256 d71db7dc5f467af7d3ffe252e7ef377351add38da3307ea469eab5e891425c6d
run-real-flagship-fight-meta-llama-4-maverick-002
meta-llama-4-maverick · model · workload v1.0.0
SHA-256 2d7075da1cdeef0195a29bdf400ad66d72758d4e136a8364872b9d1059ad7252
run-real-flagship-fight-meta-llama-4-maverick-003
meta-llama-4-maverick · model · workload v1.0.0
SHA-256 ec557a82b96ee58ef87760f6cd39475e75af16d1bd323b9171de3ca0a639b1dd
run-real-flagship-fight-mistralai-mistral-large-2512-001
mistralai-mistral-large-2512 · model · workload v1.0.0
SHA-256 330c65863e1ffc2697d71126ff1994c2d5dbdf8168c6d7d67d3621ec1acd302f
run-real-flagship-fight-mistralai-mistral-large-2512-002
mistralai-mistral-large-2512 · model · workload v1.0.0
SHA-256 c1f7579d7c92af283421e97f68a97daec909a8b253a67c3872a5a318720ef1f2
run-real-flagship-fight-mistralai-mistral-large-2512-003
mistralai-mistral-large-2512 · model · workload v1.0.0
SHA-256 bcb52e6f37aef18e353ecf8a9b828b2aa8f0dc481e58105c5813a2930ffaf0ff
run-real-flagship-fight-moonshot-kimi-k2-5-001
moonshot-kimi-k2-5 · model · workload v1.0.0
SHA-256 46794f2f9f81e82469a341940f919f09dabe0df434a0870d8b741d7e38cb1644
run-real-flagship-fight-moonshot-kimi-k2-5-002
moonshot-kimi-k2-5 · model · workload v1.0.0
SHA-256 234e0445fe757fe99029960437fd23cdcd49cf296c1e2f05832f47f7eb6166ce
run-real-flagship-fight-moonshot-kimi-k2-5-003
moonshot-kimi-k2-5 · model · workload v1.0.0
SHA-256 13b4bda820f9cad1e91318192d9b04633652a5d534ad0421560d6839f9bb8d85
run-real-flagship-fight-openai-gpt-5-1-001
openai-gpt-5-1 · model · workload v1.0.0
SHA-256 d50b191525d2baeec1b718ce0637a9bf26c071736957c6301c9e36f817e61f61
run-real-flagship-fight-openai-gpt-5-1-002
openai-gpt-5-1 · model · workload v1.0.0
SHA-256 50353bbade950472243ad700460c6724a95257a53c293ce91694f546d5829f5b
run-real-flagship-fight-openai-gpt-5-1-003
openai-gpt-5-1 · model · workload v1.0.0
SHA-256 6207d141ab1161df24c312b6617b5625757a0680aca08d905552e34416c46837
run-real-flagship-fight-qwen-qwen3-max-001
qwen-qwen3-max · model · workload v1.0.0
SHA-256 ccd63c9eebd1a4a1d1b621e94877f9bd87938ca9c342182d98b32d68d0ae0ced
run-real-flagship-fight-qwen-qwen3-max-002
qwen-qwen3-max · model · workload v1.0.0
SHA-256 2effbb613a56fb26119b62a2415e48d690a07f84e3a2011ec0275d2c89b07487
run-real-flagship-fight-qwen-qwen3-max-003
qwen-qwen3-max · model · workload v1.0.0
SHA-256 159142f5f64ac304d7d4715210a1756fc498cd816deed3b89255e237b383df8e
run-real-flagship-fight-x-ai-grok-4-5-001
x-ai-grok-4-5 · model · workload v1.0.0
SHA-256 60c088c238806876c8194082a1de76731f0e53e229ee103f38c38d6d1ab7eda9
run-real-flagship-fight-x-ai-grok-4-5-002
x-ai-grok-4-5 · model · workload v1.0.0
SHA-256 08de6e9b74bb41e63dfd3676ba28a8b57ca1ebd4f27c55dc0ca156af314a2a48
run-real-flagship-fight-x-ai-grok-4-5-003
x-ai-grok-4-5 · model · workload v1.0.0
SHA-256 76639a3f94fca23b2a5bc01d96779c348399fd1bb6aeb154b44269174a8e6fb3