{"board":{"id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","source_key":"agents","source_order":1,"label":"Agents","title":"Agent performance","source_name":"Arena","source_url":"https://arena.ai/leaderboard/agent","source_date_label":"Sep 9, 2026","total":1587202,"total_label":"sessions","metric":"Net improvement","unit":"%"},"results":[{"id":"bmr_1HN48VE9D97VA04CMAV54BC7HE","source_key":"agents:0:Claude Fable 5.1 (Max)","source_order":128,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_78CJTBD292JAC4MV3JJ6QTS1GH","model_label":"Claude Fable 5.1 (Max)","provider":"Anthropic","rank":1,"score":13.85,"margin":1.92,"sample_count":12416,"rank_spread":["1","4"],"preliminary":false,"model_url":"https://www.anthropic.com/claude-fable-and-mythos-5-1","price_label":null,"context_label":null,"license_label":"Proprietary","cost":4.48,"success_label":"+22.39 % ±3.04%","steerability_label":"-0.91 % ±3.86%","tokens_label":"55.1K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_4YBETTZKN9JP5P5FDZJX87RV68","source_key":"agents:1:GPT 6 Astra (Max)","source_order":129,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_3TPFVCVW625H40TPYZX87AZYRK","model_label":"GPT 6 Astra (Max)","provider":"OpenAI","rank":2,"score":12.39,"margin":2.6,"sample_count":9692,"rank_spread":["1","5"],"preliminary":false,"model_url":"https://openai.com/index/gpt-6-astra/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":4.09,"success_label":"+18.82 % ±3.85%","steerability_label":"-3.75 % ±5.13%","tokens_label":"33.5K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_6F3DXASCVJBZ1QWZ94GXCCYBGT","source_key":"agents:2:Claude Opus 5 (High)","source_order":130,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_1G18C9N4BPX2ZMK4CKY9P5QRG8","model_label":"Claude Opus 5 (High)","provider":"Anthropic","rank":3,"score":11.06,"margin":1.7,"sample_count":23679,"rank_spread":["1","5"],"preliminary":false,"model_url":"https://www.anthropic.com/news/claude-opus-5","price_label":null,"context_label":null,"license_label":"Proprietary","cost":2.39,"success_label":"+11.15 % ±3.56%","steerability_label":"+12.77 % ±3.30%","tokens_label":"30.1K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_0CGK4FW95VP2RPZ4HBV8HA64AM","source_key":"agents:3:Claude Opus 5 (Max)","source_order":131,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_7WZRAR36E6SW41HQB5MHG8JX68","model_label":"Claude Opus 5 (Max)","provider":"Anthropic","rank":4,"score":10.8,"margin":1.88,"sample_count":19080,"rank_spread":["1","7"],"preliminary":false,"model_url":"https://www.anthropic.com/news/claude-opus-5","price_label":null,"context_label":null,"license_label":"Proprietary","cost":3.51,"success_label":"+12.83 % ±3.79%","steerability_label":"+9.57 % ±3.72%","tokens_label":"47.7K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_4K45MQHD02671VG0S6D3ZNC5MD","source_key":"agents:4:Claude Fable 5 (High)","source_order":132,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_6XR4QYWP75FPPQ9JQDZ0BPB5WJ","model_label":"Claude Fable 5 (High)","provider":"Anthropic","rank":5,"score":9.09,"margin":1.42,"sample_count":36950,"rank_spread":["2","8"],"preliminary":false,"model_url":"https://www.anthropic.com/news/claude-fable-5-mythos-5","price_label":null,"context_label":null,"license_label":"Proprietary","cost":2.23,"success_label":"+6.59 % ±3.08%","steerability_label":"+11.20 % ±2.59%","tokens_label":"15.8K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_6BHVXQ94XG2ZFQ63WDWWHCF101","source_key":"agents:5:Claude Opus 4.8 (High)","source_order":133,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_3V0GQJEETHSJMHVESKQFH56ZAV","model_label":"Claude Opus 4.8 (High)","provider":"Anthropic","rank":6,"score":7.57,"margin":1.4,"sample_count":39316,"rank_spread":["4","11"],"preliminary":false,"model_url":"https://www.anthropic.com/news/claude-opus-4-8","price_label":null,"context_label":null,"license_label":"Proprietary","cost":1.42,"success_label":"+4.33 % ±2.95%","steerability_label":"+11.34 % ±2.65%","tokens_label":"16.8K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_3VJN2TE1P39CC2CWEBWVMHMTMM","source_key":"agents:6:GPT 5.6 Sol (xHigh)","source_order":134,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_18AB90H1CBV9JC8V06HZH8P96V","model_label":"GPT 5.6 Sol (xHigh)","provider":"OpenAI","rank":7,"score":7.51,"margin":1.42,"sample_count":30881,"rank_spread":["4","11"],"preliminary":false,"model_url":"https://openai.com/index/gpt-5-6/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":1.12,"success_label":"+5.02 % ±3.02%","steerability_label":"+8.58 % ±2.75%","tokens_label":"18.6K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_295TWGP5FNX42Z9PPZVTDVMESK","source_key":"agents:7:Kimi K3 (Max)","source_order":135,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_19NT6DQ1W4TA2DP369V18ZXB3R","model_label":"Kimi K3 (Max)","provider":"Moonshot","rank":8,"score":6.46,"margin":0.7,"sample_count":101022,"rank_spread":["6","13"],"preliminary":false,"model_url":"https://platform.kimi.ai/docs/guide/kimi-k3-quickstart","price_label":null,"context_label":null,"license_label":"Kimi K3 license","cost":0.8,"success_label":"+15.11 % ±1.43%","steerability_label":"+0.02 % ±1.39%","tokens_label":"25.2K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_0GKK4C2WTHAWCF9MFAFZA0NR0Y","source_key":"agents:8:Claude Sonnet 5 (High)","source_order":136,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_2N5R7EC75AC9P7HSA779QBCP5C","model_label":"Claude Sonnet 5 (High)","provider":"Anthropic","rank":9,"score":5.88,"margin":1.82,"sample_count":29782,"rank_spread":["5","18"],"preliminary":false,"model_url":"https://www.anthropic.com/news/claude-sonnet-5","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.86,"success_label":"+0.96 % ±3.85%","steerability_label":"+9.40 % ±3.91%","tokens_label":"22K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_6EWS7D2N625VTSTQXJW4FEKNC7","source_key":"agents:9:Hy4 preview","source_order":137,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_3AMTSW17GRTSYA1GGQMVA4SN57","model_label":"Hy4 preview","provider":"Tencent","rank":10,"score":5.23,"margin":1.31,"sample_count":19613,"rank_spread":["6","18"],"preliminary":false,"model_url":"https://hy.tencent.ai/research/hy4-preview","price_label":null,"context_label":null,"license_label":"Apache 2.0","cost":0.23,"success_label":"+11.82 % ±2.51%","steerability_label":"-2.65 % ±2.83%","tokens_label":"32.3K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_6GAV6NE9YTJKGEXX09KZGWEZ68","source_key":"agents:10:GPT 5.5 (xHigh)","source_order":138,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_20V814B07EY9TWQFESRH5HV95K","model_label":"GPT 5.5 (xHigh)","provider":"OpenAI","rank":11,"score":5.22,"margin":1.08,"sample_count":52336,"rank_spread":["6","18"],"preliminary":false,"model_url":"https://openai.com/index/introducing-gpt-5-5/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":1.05,"success_label":"-0.32 % ±2.46%","steerability_label":"+7.12 % ±2.06%","tokens_label":"17.2K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_42SK215Z9WF7E9N5AWF90XNHW7","source_key":"agents:11:GLM 5.2 (Max)","source_order":139,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_011K9D037T6WYYZTV54AZ0777P","model_label":"GLM 5.2 (Max)","provider":"Z.ai","rank":12,"score":4.56,"margin":0.74,"sample_count":73382,"rank_spread":["9","18"],"preliminary":false,"model_url":"https://huggingface.co/zai-org/GLM-5.2","price_label":null,"context_label":null,"license_label":"SiliconFlow","cost":0.64,"success_label":"+8.10 % ±1.58%","steerability_label":"+3.16 % ±1.36%","tokens_label":"42.7K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_5N03NT0HVART7DADNWBVE4YWWA","source_key":"agents:12:Muse Spark 1.3 (Max)","source_order":140,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_5MP5XJB2S59MV95HGZVFT0NREA","model_label":"Muse Spark 1.3 (Max)","provider":"Meta","rank":13,"score":4.23,"margin":1.77,"sample_count":8770,"rank_spread":["8","23"],"preliminary":false,"model_url":"https://developer.meta.com/ai/models/muse-spark/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.31,"success_label":"+8.37 % ±3.65%","steerability_label":"+1.38 % ±4.47%","tokens_label":"20.5K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_2ERJ5949WSDZM5ZX1QD7NA8A2F","source_key":"agents:13:DeepSeek V4 Pro (High) (0813)","source_order":141,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_0CGB2JTZE2P5CRDPJ658WPVFKJ","model_label":"DeepSeek V4 Pro (High) (0813)","provider":"DeepSeek","rank":14,"score":4.14,"margin":0.92,"sample_count":39047,"rank_spread":["9","20"],"preliminary":false,"model_url":"https://api-docs.deepseek.com/updates/#date-2026-08-13","price_label":null,"context_label":null,"license_label":"MIT","cost":0.23,"success_label":"+10.70 % ±1.87%","steerability_label":"-1.61 % ±1.93%","tokens_label":"32.4K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_40MA4P83BYDP4CGQA8FMRWDAHN","source_key":"agents:14:Qwen3.8 Max","source_order":142,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_5CSG0K2GZWVZ7DCB67Y5WBRRZG","model_label":"Qwen3.8 Max","provider":"Alibaba","rank":15,"score":3.97,"margin":1.02,"sample_count":23261,"rank_spread":["9","20"],"preliminary":false,"model_url":"https://qwen.ai/blog?id=qwen3.8","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.47,"success_label":"+9.52 % ±2.22%","steerability_label":"+2.60 % ±2.10%","tokens_label":"25.9K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_4X812A6Y1SW73Q9MYN7V74BPDY","source_key":"agents:15:Gemini 3.8 Flash (High)","source_order":143,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_1HB31JA6S03XN9JC13153Q8ANA","model_label":"Gemini 3.8 Flash (High)","provider":"Google","rank":16,"score":3.84,"margin":1.96,"sample_count":10105,"rank_spread":["8","24"],"preliminary":false,"model_url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":null,"success_label":"+8.53 % ±4.58%","steerability_label":"-3.37 % ±5.19%","tokens_label":"N/A","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_7G15ZS68HRAWTEGVGZXZFDGG57","source_key":"agents:16:Grok 4.5","source_order":144,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_3A9FSKJ9MAXCKQTKN6CNZ5H2XG","model_label":"Grok 4.5","provider":"SpaceXAI","rank":17,"score":3.83,"margin":1.13,"sample_count":36452,"rank_spread":["9","22"],"preliminary":false,"model_url":"https://docs.x.ai/developers/models/grok-4.5","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.5,"success_label":"+4.78 % ±2.54%","steerability_label":"+5.54 % ±2.26%","tokens_label":"13.4K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_46ZG3G4DWVG1828XTJXD6NV6NQ","source_key":"agents:17:Grok 4.6 (xHigh)","source_order":145,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_2NY5928H44DXW7CZXTXZF9J4CY","model_label":"Grok 4.6 (xHigh)","provider":"SpaceXAI","rank":18,"score":3.12,"margin":1.14,"sample_count":20911,"rank_spread":["9","24"],"preliminary":false,"model_url":"https://docs.x.ai/developers/models/grok-4.6","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.91,"success_label":"+4.37 % ±2.58%","steerability_label":"+2.42 % ±2.33%","tokens_label":"23.5K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_4JN0QJ7KMD1Z9GCCB6X0YGJYAT","source_key":"agents:18:GPT 5.5","source_order":146,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_28BAEDHSAJ1BJ2JXX035K0VGM2","model_label":"GPT 5.5","provider":"OpenAI","rank":19,"score":2.8,"margin":0.93,"sample_count":80458,"rank_spread":["13","24"],"preliminary":false,"model_url":"https://openai.com/index/introducing-gpt-5-5/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.55,"success_label":"-1.52 % ±2.12%","steerability_label":"+3.56 % ±1.69%","tokens_label":"9K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_7PXW9B0TDZV0HS6KDX0692Q8Q5","source_key":"agents:19:GLM 5.3 (Max)","source_order":147,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_0TAH1HA9VEZVF0JTRADV752SGD","model_label":"GLM 5.3 (Max)","provider":"Z.ai","rank":20,"score":2.62,"margin":0.66,"sample_count":65488,"rank_spread":["13","24"],"preliminary":false,"model_url":"https://z.ai/blog/glm-5.3","price_label":null,"context_label":null,"license_label":"MIT","cost":0.5,"success_label":"+11.24 % ±1.41%","steerability_label":"-0.88 % ±1.28%","tokens_label":"33.5K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_6QYDM1J4RE22J2CKY831F7E5EA","source_key":"agents:20:Deepseek V4 Flash (High) (20260731)","source_order":148,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_703XWMP2KH0BVDN8A4N6Q81MCK","model_label":"Deepseek V4 Flash (High) (20260731)","provider":"DeepSeek","rank":21,"score":2.17,"margin":0.77,"sample_count":63940,"rank_spread":["15","25"],"preliminary":false,"model_url":"https://api-docs.deepseek.com/updates/#date-2026-07-31","price_label":null,"context_label":null,"license_label":"MIT","cost":0.16,"success_label":"+7.06 % ±1.66%","steerability_label":"-1.45 % ±1.62%","tokens_label":"42.2K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_2F4P6FKDM9034A0Y5NGE99Z3NC","source_key":"agents:21:GLM 5.3 Flash","source_order":149,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_4R74GZA0P0TMKCS1QHKQG8W5Q1","model_label":"GLM 5.3 Flash","provider":"Z.ai","rank":22,"score":1.89,"margin":0.86,"sample_count":27007,"rank_spread":["15","26"],"preliminary":false,"model_url":"https://z.ai/blog/glm-5.3","price_label":null,"context_label":null,"license_label":"MIT","cost":0.1,"success_label":"+12.49 % ±1.84%","steerability_label":"-1.06 % ±1.80%","tokens_label":"33K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_11QA5XQTNCT1NGGZPWSG55M49W","source_key":"agents:22:GPT 5.4 (High)","source_order":150,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_4B563Q7F8V2BX72VEY6FCR2D4R","model_label":"GPT 5.4 (High)","provider":"OpenAI","rank":23,"score":1.54,"margin":0.9,"sample_count":79621,"rank_spread":["17","28"],"preliminary":false,"model_url":"https://platform.openai.com/docs/models/gpt-5.4","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.6,"success_label":"-0.33 % ±2.15%","steerability_label":"+3.40 % ±1.81%","tokens_label":"17.3K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_78BVSC3K6GC079AV3NH5WED49G","source_key":"agents:23:GPT 5.6 Terra (xHigh)","source_order":151,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_64J3RRPVSBC7QD1S8V2RC8YVX5","model_label":"GPT 5.6 Terra (xHigh)","provider":"OpenAI","rank":24,"score":1.46,"margin":1.19,"sample_count":19096,"rank_spread":["16","28"],"preliminary":false,"model_url":"https://openai.com/index/gpt-5-6/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.46,"success_label":"-4.17 % ±2.89%","steerability_label":"+6.14 % ±2.36%","tokens_label":"14.1K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_67NPSK8N324S985744JYBDWD7N","source_key":"agents:24:GPT 5.6 Luna (xHigh)","source_order":152,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_1NBSBK195QNDTMD2GRBC7EKAQ3","model_label":"GPT 5.6 Luna (xHigh)","provider":"OpenAI","rank":25,"score":0.73,"margin":1.05,"sample_count":22203,"rank_spread":["21","29"],"preliminary":false,"model_url":"https://openai.com/index/gpt-5-6/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.08,"success_label":"-0.48 % ±2.43%","steerability_label":"+0.71 % ±2.18%","tokens_label":"14.6K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_0Z19B1DKMN5RRHFJ9N674QA2RS","source_key":"agents:25:Qwen3.8 Flash Next","source_order":153,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_3BWDPQX672MQ2BGZDDHZSZ8R5K","model_label":"Qwen3.8 Flash Next","provider":"Alibaba","rank":26,"score":0.4,"margin":0.97,"sample_count":30755,"rank_spread":["22","30"],"preliminary":false,"model_url":"https://qwen.ai/blog?id=qwen3.8-flash-next","price_label":null,"context_label":null,"license_label":"Qwen-community-1.0","cost":null,"success_label":"+11.74 % ±2.01%","steerability_label":"-5.20 % ±2.23%","tokens_label":"46.2K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_00DS5486HM109WX8QF1ZMH1HYY","source_key":"agents:26:Qwen 3.8 27B","source_order":154,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_51VEZMJ2C36R9215B8EK3PCCJX","model_label":"Qwen 3.8 27B","provider":"Alibaba","rank":27,"score":-0.08,"margin":0.88,"sample_count":28536,"rank_spread":["23","30"],"preliminary":false,"model_url":"https://huggingface.co/Qwen/Qwen3.8-27B","price_label":null,"context_label":null,"license_label":"Apache 2.0","cost":0.35,"success_label":"+7.22 % ±1.97%","steerability_label":"-1.42 % ±1.81%","tokens_label":"37.4K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_7RJ1MSE0ZBERD0CRGG3WZQ4EDN","source_key":"agents:27:DeepSeek V4 Pro","source_order":155,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_7YNXR4J9KHQJ6KJ7EV1JNFB9M0","model_label":"DeepSeek V4 Pro","provider":"DeepSeek","rank":28,"score":-0.34,"margin":1.14,"sample_count":33240,"rank_spread":["23","30"],"preliminary":false,"model_url":"https://api-docs.deepseek.com/news/news260424","price_label":null,"context_label":null,"license_label":"MIT","cost":0.29,"success_label":"-0.42 % ±2.89%","steerability_label":"-1.05 % ±2.14%","tokens_label":"35.4K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_4ARHJ4T13ACY2CXQH6F5SVKZNK","source_key":"agents:28:Claude Sonnet 4.6","source_order":156,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_3RHNE8J6W9JTEN3W0BSGHA3E3J","model_label":"Claude Sonnet 4.6","provider":"Anthropic","rank":29,"score":-1.27,"margin":1.32,"sample_count":39399,"rank_spread":["25","32"],"preliminary":false,"model_url":"https://www.anthropic.com/news/claude-sonnet-4-6","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.6,"success_label":"-7.96 % ±3.35%","steerability_label":"-2.50 % ±2.58%","tokens_label":"9.6K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_15JJX9HK9RJ0HK4HMVJS3HR790","source_key":"agents:29:Muse Spark 1.2 (xHigh)","source_order":157,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_02CXGZRJ3D6QTKVPN0NQ63PMVQ","model_label":"Muse Spark 1.2 (xHigh)","provider":"Meta","rank":30,"score":-1.76,"margin":1.29,"sample_count":24050,"rank_spread":["26","32"],"preliminary":false,"model_url":"https://developer.meta.com/ai/models/muse-spark/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.3,"success_label":"+1.46 % ±3.10%","steerability_label":"-7.95 % ±2.75%","tokens_label":"20K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_5XC9R9CE6WZV7JC92QCBGBCVB5","source_key":"agents:30:Muse Spark 1.1","source_order":158,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_3HAJT00AHFPX2C2NRJXWS9SE7V","model_label":"Muse Spark 1.1","provider":"Meta","rank":31,"score":-2.97,"margin":0.65,"sample_count":95367,"rank_spread":["29","33"],"preliminary":false,"model_url":"https://ai.meta.com/blog/introducing-muse-spark-meta-model-api/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.53,"success_label":"+3.34 % ±1.55%","steerability_label":"-6.61 % ±1.32%","tokens_label":"13.1K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_45ZSTQ0F3QSZNKZQB4YZGA6PQC","source_key":"agents:31:Qwen3.7 Max","source_order":159,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_5RRDRVBN1YSV3ATS1XK88ARP5Q","model_label":"Qwen3.7 Max","provider":"Alibaba","rank":32,"score":-3.12,"margin":0.99,"sample_count":37494,"rank_spread":["29","34"],"preliminary":false,"model_url":null,"price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.21,"success_label":"-4.11 % ±2.44%","steerability_label":"-3.16 % ±1.78%","tokens_label":"12.3K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_3SY9GPPWNTDJ31SM807KWJQ6NN","source_key":"agents:32:Hy3","source_order":160,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_4R51PAJBN9HB57B6DTENR1EKYV","model_label":"Hy3","provider":"Tencent","rank":33,"score":-4.69,"margin":1.29,"sample_count":26255,"rank_spread":["31","39"],"preliminary":false,"model_url":"https://huggingface.co/tencent/Hy3","price_label":null,"context_label":null,"license_label":"Apache 2.0","cost":0.05,"success_label":"-7.63 % ±3.00%","steerability_label":"-1.81 % ±2.68%","tokens_label":"22.1K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_0AM1RB5G7X32JQNASS6F7QRCDD","source_key":"agents:33:Minimax M3","source_order":161,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_0SKWKQKCZP6MDC6EHMSYGJXPXK","model_label":"Minimax M3","provider":"MiniMax","rank":34,"score":-5.38,"margin":0.95,"sample_count":37895,"rank_spread":["33","39"],"preliminary":false,"model_url":"https://www.minimax.io/models/text/m3","price_label":null,"context_label":null,"license_label":"MiniMax Community License","cost":0.12,"success_label":"-10.21 % ±2.57%","steerability_label":"-6.82 % ±1.91%","tokens_label":"11.7K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_43VMS1DD16APEZ09K5DBZ34W4Y","source_key":"agents:34:Qwen3.7 Plus","source_order":162,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_16NQC9FHMX4B6E1S99M0JBM2F1","model_label":"Qwen3.7 Plus","provider":"Alibaba","rank":35,"score":-5.38,"margin":1.28,"sample_count":20606,"rank_spread":["32","39"],"preliminary":false,"model_url":"https://qwen.ai/blog?id=qwen3.7-plus","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.1,"success_label":"-7.75 % ±3.33%","steerability_label":"-3.89 % ±2.63%","tokens_label":"14.1K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_5NHCTZ1R4JF0WGMKSV45V2JXWY","source_key":"agents:35:Gemini 3.6 Flash (High)","source_order":163,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_5XCKJF6JMFBRFVCKGP5QKWFGQP","model_label":"Gemini 3.6 Flash (High)","provider":"Google","rank":36,"score":-5.41,"margin":1.13,"sample_count":19631,"rank_spread":["33","39"],"preliminary":false,"model_url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-6-flash-3-5-flash-lite-3-5-flash-cyber/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.36,"success_label":"-5.56 % ±2.88%","steerability_label":"-3.89 % ±2.27%","tokens_label":"25.1K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_1GA3EE3HVGY41BA2CVX2VJGG0G","source_key":"agents:36:Mimo V2.5 Pro","source_order":164,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_2720EF85ZZK33RY26PYYTKV7PX","model_label":"Mimo V2.5 Pro","provider":"Xiaomi","rank":37,"score":-5.42,"margin":0.98,"sample_count":38911,"rank_spread":["33","39"],"preliminary":false,"model_url":"https://mimo.xiaomi.com/mimo-v2-5-pro/","price_label":null,"context_label":null,"license_label":"MIT","cost":0.05,"success_label":"-7.28 % ±2.54%","steerability_label":"-3.78 % ±1.90%","tokens_label":"9.6K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_3YH39KQ9X1B59BKVKNWGSW1RP2","source_key":"agents:37:Gemini 3.1 Pro Preview","source_order":165,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_3Q2KEJEZV8RWRQ4NWC4CERXCDH","model_label":"Gemini 3.1 Pro Preview","provider":"Google","rank":38,"score":-5.6,"margin":0.96,"sample_count":85599,"rank_spread":["33","39"],"preliminary":false,"model_url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-pro/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.31,"success_label":"-5.40 % ±2.32%","steerability_label":"-2.60 % ±1.67%","tokens_label":"10.2K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_2D5M390CAZC4EP40EJV0D6CEH9","source_key":"agents:38:Inkling Small","source_order":166,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_42XGXZY23P4QPW92GNEDYQTYF2","model_label":"Inkling Small","provider":"Thinky","rank":39,"score":-7.09,"margin":1.58,"sample_count":12366,"rank_spread":["33","41"],"preliminary":false,"model_url":"https://thinkingmachines.ai/news/inkling-small/","price_label":null,"context_label":null,"license_label":"Apache 2.0","cost":0.12,"success_label":"-19.89 % ±4.54%","steerability_label":"-5.97 % ±3.71%","tokens_label":"15.7K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_1S3M7X2M5HVW8VMDEVJSYJK8C4","source_key":"agents:39:Inkling","source_order":167,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_5PV4T41CHKRQXE04FV16G7E7H7","model_label":"Inkling","provider":"Thinky","rank":40,"score":-9.12,"margin":1.06,"sample_count":41981,"rank_spread":["39","41"],"preliminary":false,"model_url":"https://thinkingmachines.ai/news/introducing-inkling/","price_label":null,"context_label":null,"license_label":"Apache 2.0","cost":0.24,"success_label":"-17.83 % ±3.18%","steerability_label":"-8.63 % ±2.45%","tokens_label":"11K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_6CAS10NCG0EB6QMV27QH6QWFGY","source_key":"agents:40:Mistral Medium 3.5","source_order":168,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_0Y2WVTEEZ1P7MGHHV5QCTY1JW7","model_label":"Mistral Medium 3.5","provider":"Mistral","rank":41,"score":-10.12,"margin":1.47,"sample_count":9879,"rank_spread":["39","41"],"preliminary":false,"model_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-5-26-04","price_label":null,"context_label":null,"license_label":"Modified MIT","cost":0.74,"success_label":"-16.44 % ±3.77%","steerability_label":"-4.37 % ±3.28%","tokens_label":"12.5K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_59XC457HQGJPP27JSJEH8ZNR2J","source_key":"agents:41:Minimax M2.7","source_order":169,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_2W74MPSMYNS5Q0EG3H727GA55X","model_label":"Minimax M2.7","provider":"MiniMax","rank":42,"score":-13.22,"margin":1.22,"sample_count":37514,"rank_spread":["42","43"],"preliminary":false,"model_url":"https://www.minimax.io/news/minimax-m27-en","price_label":null,"context_label":null,"license_label":"Modified MIT","cost":0.07,"success_label":"-16.69 % ±3.09%","steerability_label":"-6.18 % ±2.61%","tokens_label":"7.7K","win_rate":null,"win_margin":null,"seconds":null},{"id":"bmr_0G0EDYRSM92KH6QW4MGBMKKD8N","source_key":"agents:42:Gemini 3.5 Flash Lite","source_order":170,"board_id":"bmb_5QQ3V5PV95VSBEABV6N7XSAWF8","model_id":"mdl_3SMXDK0GB5YMQDM6QJ3D9MXHE5","model_label":"Gemini 3.5 Flash Lite","provider":"Google","rank":43,"score":-13.73,"margin":1.45,"sample_count":23216,"rank_spread":["42","43"],"preliminary":false,"model_url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-5/","price_label":null,"context_label":null,"license_label":"Proprietary","cost":0.08,"success_label":"-18.30 % ±3.82%","steerability_label":"-10.92 % ±2.93%","tokens_label":"9.5K","win_rate":null,"win_margin":null,"seconds":null}]}