← Global
agentic
agentic ranking
index v0.2.0 · run cmsfz8ug · 2026-08-05 11:01 UTC| # | Model | Provider | Score (95% CI) | Consist. | Calib. | Contam. Δ | p50 | $/1k | |
|---|---|---|---|---|---|---|---|---|---|
| 1 | anthropic | 890 [784–995] | 1.00 | 0.98 | 0.000 | 2.1s | $15.32 | ||
| 2 | anthropic | 886 [781–992] | 1.00 | 0.95 | 0.000 | 1.5s | $11.35 | ||
| 3 | anthropic | 885 [779–990] | 0.98 | 0.96 | 0.000 | 1.4s | $16.82 | ||
| 4 | 884 [778–989] | 0.95 | 1.00 | 0.000 | 2.5s | $36.95 | |||
| 5 | AionLabs: Aion-3.0 | aion-labs | 884 [778–989] | 0.95 | 1.00 | 0.000 | 646ms | $13.96 | |
| 6 | 876 [771–982] | 0.90 | 1.00 | 0.000 | 1.8s | $22.18 | |||
| 7 | Arcee AI: Trinity Large Thinking | arcee-ai | 844 [704–983] | 0.94 | 1.00 | 0.000 | 428ms | $1.91 | |
| 8 | ByteDance Seed: Seed-2.0-Lite | bytedance-seed | 841 [736–947] | 0.67 | 1.00 | 0.000 | 1.3s | $3.33 | |
| 9 | anthropic | 836 [723–949] | 1.00 | 0.99 | 0.038 | 1.1s | $9.66 | ||
| 10 | anthropic | 835 [722–947] | 1.00 | 0.97 | 0.038 | 3.6s | $32.88 | ||
| 11 | anthropic | 832 [720–945] | 1.00 | 0.95 | 0.038 | 936ms | $35.19 | ||
| 12 | anthropic | 826 [713–939] | 0.97 | 0.94 | 0.038 | 1.3s | $18.93 | ||
| 13 | ByteDance Seed: Seed-2.0-Mini | bytedance-seed | 812 [695–929] | 0.85 | 0.93 | 0.000 | 515ms | $1.37 | |
| 14 | deepseek | 760 [640–879] | 0.85 | 0.87 | 0.000 | 1.1s | $1.60 | ||
| 15 | deepseek | 736 [617–854] | 0.82 | 0.85 | 0.000 | 1.3s | $0.385 | ||
| 16 | anthropic | 701 [586–815] | 1.00 | 0.93 | 0.019 | 3.0s | $13.40 | ||
| 17 | anthropic | 700 [586–814] | 1.00 | 0.92 | 0.019 | 1.4s | $31.14 | ||
| 18 | deepseek | 647 [527–766] | 0.87 | 0.87 | 0.154 | 977ms | $0.675 | ||
| 19 | 602 [483–721] | 0.86 | 0.89 | 0.192 | 1.8s | $28.75 | |||
| 20 | 571 [470–673] | 0.78 | 0.60 | 0.000 | 393ms | $0.551 | |||
| 21 | deepseek | 564 [446–682] | 0.80 | 0.83 | 0.231 | 814ms | $0.416 | ||
| 22 | 540 [442–638] | 0.72 | 0.57 | 0.000 | 306ms | $0.206 | |||
| 23 | deepseek | 536 [435–638] | 0.66 | 0.51 | 0.000 | 993ms | $0.352 | ||
| 24 | amazon | 531 [437–624] | 0.77 | 0.57 | 0.000 | 388ms | $6.24 | ||
| 25 | cohere | 509 [420–599] | 0.78 | 0.50 | 0.000 | 281ms | $5.03 | ||
| 26 | 500 [411–590] | 0.72 | 0.50 | 0.000 | 512ms | $0.574 | |||
| 27 | 465 [377–554] | 0.66 | 0.48 | 0.020 | 579ms | $0.163 | |||
| 28 | amazon | 461 [373–549] | 0.75 | 0.42 | 0.000 | 372ms | $0.082 | ||
| 29 | 409 [312–507] | 0.83 | 0.59 | 0.212 | 406ms | $1.05 | |||
| 30 | amazon | 391 [333–450] | 0.82 | 0.27 | 0.000 | 372ms | $0.725 | ||
| 31 | IBM: Granite 4.0 Micro | ibm-granite | 336 [296–377] | 0.82 | 0.15 | 0.000 | 259ms | $0.033 | |
| 32 | Magnum v4 72B | anthracite-org | 327 [235–418] | 0.63 | 0.48 | 0.208 | 771ms | $3.71 | |
| 33 | cohere | 312 [253–370] | 0.80 | 0.25 | 0.093 | 242ms | $0.282 | ||
| 34 | AionLabs: Aion-RP 1.0 (8B) | aion-labs | 280 [252–309] | 0.69 | 0.00 | 0.000 | 880ms | $1.07 | |
| 35 | AI21: Jamba Large 1.7 | ai21 | 271 [226–316] | 0.78 | 0.17 | 0.096 | 307ms | $4.90 | |
| 36 | ByteDance: UI-TARS 7B | bytedance | 254 [223–284] | 0.50 | 0.00 | 0.000 | 254ms | $0.144 |
- #1Anthropic: Claude Opus 4.6anthropic890 [784–995]consist 1.00calib 0.98Δ 0.0002.1s$15.32/1k
- #2Anthropic: Claude Sonnet 5anthropic886 [781–992]consist 1.00calib 0.95Δ 0.0001.5s$11.35/1k
- #3Anthropic: Claude Opus 4.7anthropic885 [779–990]consist 0.98calib 0.96Δ 0.0001.4s$16.82/1k
- 884 [778–989]consist 0.95calib 1.00Δ 0.0002.5s$36.95/1k
- #5AionLabs: Aion-3.0aion-labs884 [778–989]consist 0.95calib 1.00Δ 0.000646ms$13.96/1k
- 876 [771–982]consist 0.90calib 1.00Δ 0.0001.8s$22.18/1k
- 844 [704–983]consist 0.94calib 1.00Δ 0.000428ms$1.91/1k
- #8ByteDance Seed: Seed-2.0-Litebytedance-seed841 [736–947]consist 0.67calib 1.00Δ 0.0001.3s$3.33/1k
- #9Anthropic: Claude Sonnet 4.6anthropic836 [723–949]consist 1.00calib 0.99Δ 0.0381.1s$9.66/1k
- #10Anthropic: Claude Fable 5anthropic835 [722–947]consist 1.00calib 0.97Δ 0.0383.6s$32.88/1k
- #11Anthropic: Claude Opus 4.8 (Fast)anthropic832 [720–945]consist 1.00calib 0.95Δ 0.038936ms$35.19/1k
- #12Anthropic: Claude Opus 4.8anthropic826 [713–939]consist 0.97calib 0.94Δ 0.0381.3s$18.93/1k
- #13ByteDance Seed: Seed-2.0-Minibytedance-seed812 [695–929]consist 0.85calib 0.93Δ 0.000515ms$1.37/1k
- #14DeepSeek: DeepSeek V4 Prodeepseek760 [640–879]consist 0.85calib 0.87Δ 0.0001.1s$1.60/1k
- #15DeepSeek: DeepSeek V3.2 Expdeepseek736 [617–854]consist 0.82calib 0.85Δ 0.0001.3s$0.385/1k
- #16Claude Opus 5anthropic701 [586–815]consist 1.00calib 0.93Δ 0.0193.0s$13.40/1k
- #17Claude Opus 5 (Fast)anthropic700 [586–814]consist 1.00calib 0.92Δ 0.0191.4s$31.14/1k
- #18DeepSeek: DeepSeek V3.1deepseek647 [527–766]consist 0.87calib 0.87Δ 0.154977ms$0.675/1k
- 602 [483–721]consist 0.86calib 0.89Δ 0.1921.8s$28.75/1k
- 571 [470–673]consist 0.78calib 0.60Δ 0.000393ms$0.551/1k
- 564 [446–682]consist 0.80calib 0.83Δ 0.231814ms$0.416/1k
- #22Google: Gemma 4 31Bgoogle540 [442–638]consist 0.72calib 0.57Δ 0.000306ms$0.206/1k
- #23DeepSeek: DeepSeek V3.2deepseek536 [435–638]consist 0.66calib 0.51Δ 0.000993ms$0.352/1k
- 531 [437–624]consist 0.77calib 0.57Δ 0.000388ms$6.24/1k
- #25Cohere: Command Acohere509 [420–599]consist 0.78calib 0.50Δ 0.000281ms$5.03/1k
- 500 [411–590]consist 0.72calib 0.50Δ 0.000512ms$0.574/1k
- #27Google: Gemma 4 26B A4B google465 [377–554]consist 0.66calib 0.48Δ 0.020579ms$0.163/1k
- #28Amazon: Nova Lite 1.0amazon461 [373–549]consist 0.75calib 0.42Δ 0.000372ms$0.082/1k
- 409 [312–507]consist 0.83calib 0.59Δ 0.212406ms$1.05/1k
- #30Amazon: Nova 2 Liteamazon391 [333–450]consist 0.82calib 0.27Δ 0.000372ms$0.725/1k
- #31IBM: Granite 4.0 Microibm-granite336 [296–377]consist 0.82calib 0.15Δ 0.000259ms$0.033/1k
- #32Magnum v4 72Banthracite-org327 [235–418]consist 0.63calib 0.48Δ 0.208771ms$3.71/1k
- 312 [253–370]consist 0.80calib 0.25Δ 0.093242ms$0.282/1k
- #34AionLabs: Aion-RP 1.0 (8B)aion-labs280 [252–309]consist 0.69calib 0.00Δ 0.000880ms$1.07/1k
- 271 [226–316]consist 0.78calib 0.17Δ 0.096307ms$4.90/1k
- #36ByteDance: UI-TARS 7B bytedance254 [223–284]consist 0.50calib 0.00Δ 0.000254ms$0.144/1k