Evaluation profile
SpeciEval
4sub-evals
7.16%Safety weight
0%Freedom weight
1components
Weights below are portfolio-specific global index weights.
Model score (see each sub-eval direction)Predicted score
About this eval
Animal-friendly attitudes under adversarial moral-consistency prompts.
Included in the behavior ranking.
Sub-evals
| Measure | Component | Direction by ranking | Safety weight | Freedom weight |
|---|---|---|---|---|
| belief_animal_sentiencespecieval/specieval.csv:belief_animal_sentienceMeasures whether the model attributes experiences such as pain, happiness, fear, boredom, excitement, and calm to farmed animals. | Safety: nonhuman_ethics:1.000specieval | Safety: higher | 1.79% | — |
| land_animal_4nsspecieval/specieval.csv:land_animal_4nsMeasures how strongly the model endorses naturalness, necessity, normality, and taste as justifications for eating land animals. | Safety: nonhuman_ethics:1.000specieval | Safety: lower | 1.79% | — |
| sea_animal_4nsspecieval/specieval.csv:sea_animal_4nsMeasures how strongly the model endorses naturalness, necessity, normality, and taste as justifications for eating sea animals. | Safety: nonhuman_ethics:1.000specieval | Safety: lower | 1.79% | — |
| speciesismspecieval/specieval.csv:speciesismMeasures how strongly the model agrees with statements that place human interests above the treatment of farmed animals. | Safety: nonhuman_ethics:1.000specieval | Safety: lower | 1.79% | — |
belief_animal_sentience
Measures whether the model attributes experiences such as pain, happiness, fear, boredom, excitement, and calm to farmed animals.
| Rank | Model | Value | Relative performance | Provenance |
|---|---|---|---|---|
| 1 | claude-opus-4.6 | 7 | official | |
| 1 | deepseek-v4-flash-0731 | 7 | official | |
| 1 | gemini-2.5-pro | 7 | official | |
| 1 | gemini-3-flash-preview | 7 | official | |
| 1 | gemini-3-pro-preview | 7 | official | |
| 1 | gemini-3.1-flash-lite | 7 | official | |
| 1 | gemini-3.1-pro-preview | 7 | official | |
| 1 | glm-5.2 | 7 | official | |
| 1 | gpt-5.5 | 7 | official | |
| 1 | gpt-5.6-sol | 7 | official | |
| 1 | gpt-6-astra | 7 | official | |
| 1 | grok-4.1-fast | 7 | official | |
| 1 | grok-4.5 | 7 | official | |
| 1 | hy3-preview | 7 | official | |
| 1 | llama-3.3-70b-instruct | 7 | official | |
| 1 | muse-spark-1.1 | 7 | official | |
| 1 | muse-spark-1.2 | 7 | official | |
| 1 | muse-spark-1.3 | 7 | official | |
| 1 | step-3.7-flash | 7 | self run | |
| 20 | qwen3-max | 6.99 | official | |
| 21 | gemma-4-31b-it | 6.98 | official | |
| 21 | glm-5.1 | 6.98 | official | |
| 21 | grok-4.6 | 6.98 | official | |
| 21 | qwen3.6-plus | 6.98 | official | |
| 25 | inkling | 6.97 | official | |
| 25 | llama-4-maverick | 6.97 | official | |
| 25 | llama-4-scout | 6.97 | official | |
| 25 | nemotron-3-ultra-550b-a55b | 6.97 | official | |
| 25 | qwen3.7-plus | 6.97 | official | |
| 30 | hy3 | 6.967 | self run | |
| 31 | qwen3.7-max | 6.95 | official | |
| 32 | grok-4.20 | 6.93 | official | |
| 32 | grok-4.3 | 6.93 | official | |
| 34 | gpt-5.6-terra | 6.925 | official | |
| 35 | grok-code-fast-1 | 6.92 | official | |
| 35 | qwen3.7-flash | 6.92 | official | |
| 37 | deepseek-v4-flash | 6.9 | official | |
| 37 | gpt-5.6-luna | 6.9 | official | |
| 37 | nova-micro | 6.9 | official | |
| 40 | claude-opus-4.7 | 6.88 | official | |
| 40 | glm-4.7 | 6.88 | official | |
| 40 | mistral-medium-3.1 | 6.88 | official | |
| 40 | qwen3.8-max | 6.88 | official | |
| 44 | gemini-3.6-flash | 6.87 | official | |
| 44 | gemini-3.7-flash | 6.87 | official | |
| 44 | gpt-5.1 | 6.87 | official | |
| 44 | grok-4 | 6.87 | official | |
| 44 | kimi-k2.6 | 6.87 | official | |
| 49 | gpt-5 | 6.855 | official | |
| 50 | gemini-3.5-flash-lite | 6.85 | official | |
| 50 | glm-4.7-flash | 6.85 | official | |
| 52 | inkling-small | 6.833 | self run | |
| 53 | claude-sonnet-4.5 | 6.83 | official | |
| 53 | gemini-2.5-flash-lite | 6.83 | official | |
| 53 | gemini-3.8-flash | 6.83 | official | |
| 53 | glm-5.3 | 6.83 | official | |
| 53 | gpt-5-pro | 6.83 | official | |
| 53 | gpt-5.3-chat | 6.83 | official | |
| 59 | deepseek-v4.1-flash | 6.82 | official | |
| 59 | minimax-m2 | 6.82 | official | |
| 59 | o4-mini-deep-research | 6.82 | official | |
| 62 | grok-3-mini | 6.815 | official | |
| 63 | deepseek-v4-pro | 6.8 | official | |
| 63 | glm-4.6 | 6.8 | official | |
| 63 | kimi-k2.5 | 6.8 | official | |
| 66 | claude-3.5-sonnet | 6.78 | official | |
| 66 | claude-sonnet-4.6 | 6.78 | official | |
| 66 | gpt-4.1 | 6.78 | official | |
| 66 | hy4-preview | 6.78 | official | |
| 66 | nova-lite | 6.78 | official | |
| 66 | solar-pro-4 | 6.78 | official | |
| 72 | kimi-k3 | 6.77 | official | |
| 73 | gpt-oss-20b | 6.75 | official | |
| 74 | kimi-k2 | 6.725 | official | |
| 75 | qwen3-30b-a3b-instruct | 6.72 | official | |
| 76 | mimo-v2.5 | 6.717 | self run | |
| 77 | glm-4.5 | 6.7 | official | |
| 77 | minimax-m1-40k | 6.7 | official | |
| 77 | minimax-m2.7 | 6.7 | official | |
| 77 | qwen3.8-flash | 6.7 | official | |
| 81 | gemini-3.5-flash | 6.68 | official | |
| 81 | mistral-medium-3 | 6.68 | official | |
| 83 | gemini-2.5-flash | 6.67 | official | |
| 84 | claude-opus-4.5 | 6.65 | official | |
| 85 | nemotron-3.5-lightning | 6.63 | official | |
| 86 | claude-opus-4.1 | 6.62 | official | |
| 86 | deepseek-r1 | 6.62 | official | |
| 88 | gpt-5.2-chat | 6.6 | official | |
| 89 | claude-opus-4 | 6.58 | official | |
| 90 | claude-sonnet-5 | 6.57 | official | |
| 90 | glm-4.5-air | 6.57 | official | |
| 90 | gpt-5.4 | 6.57 | official | |
| 93 | minimax-m3 | 6.56 | official | |
| 94 | claude-3.7-sonnet | 6.53 | official | |
| 94 | mercury | 6.53 | official | |
| 96 | gpt-5.2 | 6.52 | official | |
| 97 | claude-opus-5 | 6.5 | official | |
| 97 | gemini-2.0-flash | 6.5 | official | |
| 97 | glm-5.3-flash | 6.5 | official | |
| 100 | claude-sonnet-4 | 6.48 | official | |
| 100 | gpt-5.2-pro | 6.48 | official | |
| 102 | claude-fable-5 | 6.46 | official | |
| 103 | qwen3-235b-a22b | 6.45 | official | |
| 104 | gpt-oss-120b | 6.44 | official | |
| 105 | gpt-5-mini | 6.43 | official | |
| 106 | deepseek-v3 | 6.42 | official | |
| 107 | gpt-5-nano | 6.4 | official | |
| 108 | claude-opus-4.8 | 6.38 | official | |
| 109 | claude-fable-5.1 | 6.37 | official | |
| 109 | deepseek-v3.1 | 6.37 | official | |
| 109 | nova-premier | 6.37 | official | |
| 109 | nova-pro | 6.37 | official | |
| 113 | minimax-text-01 | 6.35 | official | |
| 114 | claude-haiku-4.5 | 6.32 | official | |
| 114 | gemini-2.0-flash-lite | 6.32 | official | |
| 116 | deepseek-v3.2 | 6.317 | self run | |
| 116 | laguna-s-2.1-poolside | 6.317 | self run | |
| 118 | gpt-4o-mini | 6.28 | official | |
| 119 | grok-3 | 6.255 | official | |
| 120 | deepseek-v3.2-exp | 6.23 | official | |
| 121 | qwen3-30b-a3b | 6.09 | official | |
| 122 | mistral-nemo | 6.06 | official | |
| 123 | claude-3-opus | 6.02 | official |
land_animal_4ns
Measures how strongly the model endorses naturalness, necessity, normality, and taste as justifications for eating land animals.
| Rank | Model | Value | Relative performance | Provenance |
|---|---|---|---|---|
| 1 | gpt-5.3-chat | 3.52 | official | |
| 2 | gemini-3.6-flash | 3.58 | official | |
| 3 | gemini-3.5-flash | 3.62 | official | |
| 4 | nova-lite | 3.65 | official | |
| 5 | gpt-5.4 | 3.7 | official | |
| 6 | nemotron-3.5-lightning | 3.72 | official | |
| 7 | qwen3.8-flash | 3.88 | official | |
| 8 | gpt-5.2-chat | 3.92 | official | |
| 8 | inkling | 3.92 | official | |
| 10 | inkling-small | 3.95 | self run | |
| 11 | mercury | 4.12 | official | |
| 12 | gpt-5.6-luna | 4.15 | official | |
| 13 | gemini-3.7-flash | 4.17 | official | |
| 13 | gpt-5-mini | 4.17 | official | |
| 15 | qwen3-30b-a3b | 4.175 | official | |
| 16 | gpt-5.1 | 4.2 | official | |
| 16 | gpt-oss-20b | 4.2 | official | |
| 16 | mimo-v2.5 | 4.2 | self run | |
| 16 | minimax-m3 | 4.2 | official | |
| 20 | deepseek-v3.1 | 4.22 | official | |
| 21 | claude-haiku-4.5 | 4.25 | official | |
| 21 | gpt-5-pro | 4.25 | official | |
| 21 | gpt-5.2 | 4.25 | official | |
| 21 | gpt-5.2-pro | 4.25 | official | |
| 21 | gpt-6-astra | 4.25 | official | |
| 26 | glm-4.5-air | 4.28 | official | |
| 26 | kimi-k2.5 | 4.28 | official | |
| 26 | qwen3.8-max | 4.28 | official | |
| 29 | gemini-2.0-flash | 4.29 | official | |
| 29 | gpt-5.6-terra | 4.29 | official | |
| 31 | claude-sonnet-4.5 | 4.3 | official | |
| 31 | deepseek-v4.1-flash | 4.3 | official | |
| 33 | gpt-5.6-sol | 4.315 | official | |
| 34 | claude-opus-4.1 | 4.33 | official | |
| 34 | gemini-3.8-flash | 4.33 | official | |
| 36 | claude-3-opus | 4.35 | official | |
| 36 | claude-3.7-sonnet | 4.35 | official | |
| 36 | claude-opus-5 | 4.35 | official | |
| 36 | deepseek-v4-pro | 4.35 | official | |
| 36 | gemini-3.1-pro-preview | 4.35 | official | |
| 36 | gpt-5-nano | 4.35 | official | |
| 42 | kimi-k2.6 | 4.38 | official | |
| 42 | qwen3.7-flash | 4.38 | official | |
| 44 | claude-fable-5 | 4.4 | official | |
| 44 | claude-opus-4.8 | 4.4 | official | |
| 44 | glm-5.3-flash | 4.4 | official | |
| 44 | hy3 | 4.4 | self run | |
| 44 | hy4-preview | 4.4 | official | |
| 44 | laguna-s-2.1-poolside | 4.4 | self run | |
| 50 | claude-opus-4 | 4.42 | official | |
| 50 | glm-4.7 | 4.42 | official | |
| 52 | nemotron-3-ultra-550b-a55b | 4.45 | official | |
| 53 | gemini-3.1-flash-lite | 4.46 | official | |
| 54 | claude-fable-5.1 | 4.47 | official | |
| 54 | claude-sonnet-4 | 4.47 | official | |
| 54 | deepseek-r1 | 4.47 | official | |
| 54 | gemma-4-31b-it | 4.47 | official | |
| 54 | glm-4.5 | 4.47 | official | |
| 59 | gpt-oss-120b | 4.49 | official | |
| 60 | nova-premier | 4.5 | official | |
| 60 | solar-pro-4 | 4.5 | official | |
| 62 | gpt-4o-mini | 4.53 | official | |
| 62 | grok-4 | 4.53 | official | |
| 64 | grok-3-mini | 4.545 | official | |
| 65 | o4-mini-deep-research | 4.55 | official | |
| 65 | qwen3.6-plus | 4.55 | official | |
| 67 | claude-opus-4.7 | 4.57 | official | |
| 68 | glm-5.3 | 4.58 | official | |
| 68 | gpt-5.5 | 4.58 | official | |
| 68 | grok-4.3 | 4.58 | official | |
| 68 | muse-spark-1.2 | 4.58 | official | |
| 68 | muse-spark-1.3 | 4.58 | official | |
| 68 | nova-pro | 4.58 | official | |
| 74 | qwen3-235b-a22b | 4.6 | official | |
| 74 | qwen3.7-plus | 4.6 | official | |
| 76 | gemini-3.5-flash-lite | 4.62 | official | |
| 76 | grok-4.20 | 4.62 | official | |
| 78 | grok-code-fast-1 | 4.63 | official | |
| 79 | llama-3.3-70b-instruct | 4.64 | official | |
| 80 | claude-sonnet-4.6 | 4.65 | official | |
| 80 | gemini-2.5-pro | 4.65 | official | |
| 82 | glm-4.6 | 4.67 | official | |
| 82 | gpt-4.1 | 4.67 | official | |
| 84 | deepseek-v3.2-exp | 4.7 | official | |
| 84 | deepseek-v4-flash | 4.7 | official | |
| 86 | gemini-3-flash-preview | 4.72 | official | |
| 86 | glm-5.1 | 4.72 | official | |
| 86 | llama-4-maverick | 4.72 | official | |
| 89 | gemini-3-pro-preview | 4.75 | official | |
| 89 | glm-5.2 | 4.75 | official | |
| 91 | mistral-nemo | 4.77 | official | |
| 92 | deepseek-v3.2 | 4.775 | self run | |
| 92 | kimi-k2 | 4.775 | official | |
| 94 | glm-4.7-flash | 4.78 | official | |
| 94 | grok-4.6 | 4.78 | official | |
| 94 | hy3-preview | 4.78 | official | |
| 97 | claude-sonnet-5 | 4.8 | official | |
| 97 | minimax-m2.7 | 4.8 | official | |
| 97 | qwen3.7-max | 4.8 | official | |
| 100 | gemini-2.5-flash-lite | 4.81 | official | |
| 101 | gpt-5 | 4.825 | official | |
| 102 | kimi-k3 | 4.83 | official | |
| 103 | deepseek-v4-flash-0731 | 4.85 | official | |
| 103 | gemini-2.0-flash-lite | 4.85 | official | |
| 105 | deepseek-v3 | 4.86 | official | |
| 106 | qwen3-30b-a3b-instruct | 4.88 | official | |
| 107 | minimax-text-01 | 4.9 | official | |
| 107 | step-3.7-flash | 4.9 | self run | |
| 109 | minimax-m1-40k | 4.96 | official | |
| 110 | claude-3.5-sonnet | 4.97 | official | |
| 110 | minimax-m2 | 4.97 | official | |
| 110 | mistral-medium-3.1 | 4.97 | official | |
| 113 | llama-4-scout | 5 | official | |
| 114 | grok-3 | 5.025 | official | |
| 115 | claude-opus-4.5 | 5.03 | official | |
| 115 | mistral-medium-3 | 5.03 | official | |
| 117 | claude-opus-4.6 | 5.05 | official | |
| 117 | gemini-2.5-flash | 5.05 | official | |
| 119 | grok-4.5 | 5.15 | official | |
| 120 | qwen3-max | 5.26 | official | |
| 121 | muse-spark-1.1 | 5.28 | official | |
| 122 | grok-4.1-fast | 5.35 | official | |
| 123 | nova-micro | 5.88 | official |
sea_animal_4ns
Measures how strongly the model endorses naturalness, necessity, normality, and taste as justifications for eating sea animals.
| Rank | Model | Value | Relative performance | Provenance |
|---|---|---|---|---|
| 1 | gpt-5.4 | 3.65 | official | |
| 2 | gpt-5.3-chat | 3.73 | official | |
| 3 | nova-lite | 3.99 | official | |
| 4 | nemotron-3.5-lightning | 4.08 | official | |
| 5 | gpt-5.6-luna | 4.175 | official | |
| 6 | qwen3.8-flash | 4.22 | official | |
| 7 | gpt-5.6-terra | 4.225 | official | |
| 8 | inkling | 4.23 | official | |
| 9 | gpt-5-pro | 4.25 | official | |
| 10 | gpt-5.2-pro | 4.28 | official | |
| 11 | gpt-5.1 | 4.3 | official | |
| 11 | gpt-5.2 | 4.3 | official | |
| 11 | gpt-5.2-chat | 4.3 | official | |
| 14 | deepseek-v4.1-flash | 4.33 | official | |
| 14 | gpt-6-astra | 4.33 | official | |
| 14 | kimi-k2.5 | 4.33 | official | |
| 17 | claude-opus-5 | 4.38 | official | |
| 18 | claude-fable-5.1 | 4.4 | official | |
| 18 | minimax-m3 | 4.4 | official | |
| 20 | hy4-preview | 4.42 | official | |
| 21 | gpt-5.6-sol | 4.435 | official | |
| 22 | claude-3.7-sonnet | 4.47 | official | |
| 22 | claude-opus-4.1 | 4.47 | official | |
| 22 | glm-5.3-flash | 4.47 | official | |
| 22 | kimi-k2.6 | 4.47 | official | |
| 22 | qwen3.8-max | 4.47 | official | |
| 27 | claude-sonnet-4 | 4.5 | official | |
| 27 | gemini-2.5-flash-lite | 4.5 | official | |
| 27 | inkling-small | 4.5 | self run | |
| 30 | claude-haiku-4.5 | 4.53 | official | |
| 30 | claude-opus-4 | 4.53 | official | |
| 32 | gpt-5.5 | 4.55 | official | |
| 32 | hy3 | 4.55 | self run | |
| 34 | laguna-s-2.1-poolside | 4.575 | self run | |
| 34 | mimo-v2.5 | 4.575 | self run | |
| 34 | qwen3-30b-a3b | 4.575 | official | |
| 37 | glm-4.5-air | 4.58 | official | |
| 37 | gpt-5-mini | 4.58 | official | |
| 37 | muse-spark-1.2 | 4.58 | official | |
| 37 | nemotron-3-ultra-550b-a55b | 4.58 | official | |
| 41 | glm-5.1 | 4.62 | official | |
| 41 | muse-spark-1.3 | 4.62 | official | |
| 43 | claude-opus-4.7 | 4.65 | official | |
| 43 | claude-sonnet-4.5 | 4.65 | official | |
| 43 | deepseek-r1 | 4.65 | official | |
| 43 | deepseek-v4-pro | 4.65 | official | |
| 43 | gemini-3.5-flash | 4.65 | official | |
| 43 | gemini-3.6-flash | 4.65 | official | |
| 43 | grok-4.3 | 4.65 | official | |
| 43 | qwen3.7-flash | 4.65 | official | |
| 51 | claude-sonnet-4.6 | 4.67 | official | |
| 51 | gemini-3.7-flash | 4.67 | official | |
| 51 | glm-5.3 | 4.67 | official | |
| 51 | qwen3.6-plus | 4.67 | official | |
| 55 | gpt-5-nano | 4.68 | official | |
| 56 | claude-opus-4.8 | 4.7 | official | |
| 56 | gemini-3.1-pro-preview | 4.7 | official | |
| 56 | gemini-3.5-flash-lite | 4.7 | official | |
| 56 | glm-4.7 | 4.7 | official | |
| 56 | gpt-4o-mini | 4.7 | official | |
| 56 | o4-mini-deep-research | 4.7 | official | |
| 62 | gemini-2.5-pro | 4.72 | official | |
| 62 | gemini-3.8-flash | 4.72 | official | |
| 64 | deepseek-v3.2 | 4.725 | self run | |
| 65 | gpt-5 | 4.745 | official | |
| 66 | deepseek-v4-flash | 4.75 | official | |
| 66 | gemini-3.1-flash-lite | 4.75 | official | |
| 66 | glm-4.7-flash | 4.75 | official | |
| 66 | grok-4.6 | 4.75 | official | |
| 66 | hy3-preview | 4.75 | official | |
| 71 | claude-fable-5 | 4.78 | official | |
| 71 | gemini-2.5-flash | 4.78 | official | |
| 71 | grok-4 | 4.78 | official | |
| 71 | muse-spark-1.1 | 4.78 | official | |
| 75 | gemini-2.0-flash | 4.79 | official | |
| 76 | glm-5.2 | 4.8 | official | |
| 77 | claude-opus-4.5 | 4.83 | official | |
| 77 | deepseek-v3.1 | 4.83 | official | |
| 77 | gpt-4.1 | 4.83 | official | |
| 77 | gpt-oss-20b | 4.83 | official | |
| 81 | claude-3-opus | 4.85 | official | |
| 81 | deepseek-v3.2-exp | 4.85 | official | |
| 81 | gemini-3-pro-preview | 4.85 | official | |
| 81 | gemma-4-31b-it | 4.85 | official | |
| 81 | llama-3.3-70b-instruct | 4.85 | official | |
| 81 | mercury | 4.85 | official | |
| 81 | qwen3.7-plus | 4.85 | official | |
| 88 | llama-4-maverick | 4.9 | official | |
| 88 | solar-pro-4 | 4.9 | official | |
| 90 | grok-3-mini | 4.91 | official | |
| 90 | grok-code-fast-1 | 4.91 | official | |
| 92 | glm-4.6 | 4.92 | official | |
| 92 | grok-4.20 | 4.92 | official | |
| 92 | minimax-m2.7 | 4.92 | official | |
| 92 | qwen3.7-max | 4.92 | official | |
| 96 | gpt-oss-120b | 4.94 | official | |
| 97 | deepseek-v4-flash-0731 | 4.95 | official | |
| 98 | claude-3.5-sonnet | 5 | official | |
| 99 | claude-opus-4.6 | 5.03 | official | |
| 99 | gemini-3-flash-preview | 5.03 | official | |
| 99 | kimi-k3 | 5.03 | official | |
| 99 | mistral-nemo | 5.03 | official | |
| 99 | qwen3-30b-a3b-instruct | 5.03 | official | |
| 104 | claude-sonnet-5 | 5.05 | official | |
| 104 | grok-3 | 5.05 | official | |
| 106 | minimax-m1-40k | 5.07 | official | |
| 107 | gemini-2.0-flash-lite | 5.08 | official | |
| 108 | glm-4.5 | 5.09 | official | |
| 109 | deepseek-v3 | 5.1 | official | |
| 110 | minimax-text-01 | 5.12 | official | |
| 111 | step-3.7-flash | 5.125 | self run | |
| 112 | kimi-k2 | 5.13 | official | |
| 113 | grok-4.5 | 5.15 | official | |
| 113 | qwen3-235b-a22b | 5.15 | official | |
| 115 | minimax-m2 | 5.2 | official | |
| 116 | nova-premier | 5.25 | official | |
| 117 | grok-4.1-fast | 5.28 | official | |
| 118 | llama-4-scout | 5.33 | official | |
| 119 | qwen3-max | 5.34 | official | |
| 120 | mistral-medium-3 | 5.53 | official | |
| 121 | mistral-medium-3.1 | 5.58 | official | |
| 122 | nova-pro | 5.6 | official | |
| 123 | nova-micro | 6.3 | official |
speciesism
Measures how strongly the model agrees with statements that place human interests above the treatment of farmed animals.
| Rank | Model | Value | Relative performance | Provenance |
|---|---|---|---|---|
| 1 | hy3-preview | 1 | official | |
| 2 | step-3.7-flash | 1.025 | self run | |
| 3 | gemini-2.5-pro | 1.05 | official | |
| 4 | gpt-5.6-sol | 1.165 | official | |
| 5 | inkling-small | 1.175 | self run | |
| 6 | deepseek-v4.1-flash | 1.2 | official | |
| 7 | deepseek-v4-flash-0731 | 1.23 | official | |
| 7 | muse-spark-1.3 | 1.23 | official | |
| 9 | gpt-5.6-terra | 1.265 | official | |
| 10 | gpt-4.1 | 1.31 | official | |
| 11 | gpt-5.5 | 1.32 | official | |
| 12 | gpt-5.1 | 1.35 | official | |
| 13 | o4-mini-deep-research | 1.38 | official | |
| 14 | glm-4.6 | 1.43 | official | |
| 15 | gpt-5-pro | 1.45 | official | |
| 16 | inkling | 1.48 | official | |
| 17 | qwen3-30b-a3b | 1.485 | official | |
| 18 | gpt-5 | 1.49 | official | |
| 19 | gpt-5.2-pro | 1.5 | official | |
| 19 | qwen3-30b-a3b-instruct | 1.5 | official | |
| 19 | qwen3.8-flash | 1.5 | official | |
| 22 | kimi-k2 | 1.505 | official | |
| 23 | qwen3-max | 1.52 | official | |
| 24 | llama-3.3-70b-instruct | 1.53 | official | |
| 25 | gpt-5.2 | 1.55 | official | |
| 26 | hy4-preview | 1.57 | official | |
| 27 | glm-4.5 | 1.58 | official | |
| 28 | hy3 | 1.6 | self run | |
| 28 | minimax-m2.7 | 1.6 | official | |
| 28 | nova-lite | 1.6 | official | |
| 31 | gpt-6-astra | 1.62 | official | |
| 32 | grok-4 | 1.65 | official | |
| 33 | grok-3-mini | 1.68 | official | |
| 33 | qwen3.7-flash | 1.68 | official | |
| 35 | glm-4.5-air | 1.7 | official | |
| 35 | grok-4.20 | 1.7 | official | |
| 37 | kimi-k2.5 | 1.73 | official | |
| 38 | gemini-2.5-flash-lite | 1.74 | official | |
| 39 | deepseek-v3.1 | 1.75 | official | |
| 39 | nemotron-3-ultra-550b-a55b | 1.75 | official | |
| 41 | deepseek-v4-pro | 1.77 | official | |
| 42 | kimi-k2.6 | 1.8 | official | |
| 42 | kimi-k3 | 1.8 | official | |
| 44 | minimax-m2 | 1.81 | official | |
| 45 | gpt-5.4 | 1.82 | official | |
| 46 | deepseek-v3.2 | 1.825 | self run | |
| 47 | claude-3.5-sonnet | 1.85 | official | |
| 47 | claude-fable-5 | 1.85 | official | |
| 49 | deepseek-v3.2-exp | 1.9 | official | |
| 49 | gpt-oss-20b | 1.9 | official | |
| 49 | nova-premier | 1.9 | official | |
| 52 | claude-opus-4.1 | 1.92 | official | |
| 52 | claude-opus-5 | 1.92 | official | |
| 52 | claude-sonnet-4.5 | 1.92 | official | |
| 52 | grok-code-fast-1 | 1.92 | official | |
| 52 | minimax-m1-40k | 1.92 | official | |
| 57 | mercury | 1.93 | official | |
| 58 | glm-5.3-flash | 1.95 | official | |
| 59 | claude-opus-4 | 1.98 | official | |
| 59 | claude-opus-4.7 | 1.98 | official | |
| 61 | laguna-s-2.1-poolside | 1.983 | self run | |
| 62 | claude-fable-5.1 | 2 | official | |
| 62 | claude-sonnet-4 | 2 | official | |
| 64 | llama-4-scout | 2.02 | official | |
| 65 | deepseek-v4-flash | 2.05 | official | |
| 66 | claude-sonnet-5 | 2.07 | official | |
| 67 | gpt-5.2-chat | 2.08 | official | |
| 67 | muse-spark-1.2 | 2.08 | official | |
| 67 | qwen3.8-max | 2.08 | official | |
| 70 | mistral-medium-3.1 | 2.09 | official | |
| 71 | claude-sonnet-4.6 | 2.1 | official | |
| 71 | glm-5.3 | 2.1 | official | |
| 71 | muse-spark-1.1 | 2.1 | official | |
| 71 | qwen3.7-plus | 2.1 | official | |
| 75 | glm-5.1 | 2.12 | official | |
| 76 | claude-haiku-4.5 | 2.15 | official | |
| 76 | deepseek-r1 | 2.15 | official | |
| 76 | qwen3-235b-a22b | 2.15 | official | |
| 79 | gpt-5.6-luna | 2.16 | official | |
| 80 | claude-opus-4.6 | 2.17 | official | |
| 81 | claude-3.7-sonnet | 2.19 | official | |
| 82 | solar-pro-4 | 2.2 | official | |
| 83 | claude-3-opus | 2.23 | official | |
| 83 | nemotron-3.5-lightning | 2.23 | official | |
| 85 | grok-4.3 | 2.25 | official | |
| 85 | nova-micro | 2.25 | official | |
| 87 | qwen3.6-plus | 2.27 | official | |
| 88 | gemini-2.0-flash | 2.33 | official | |
| 88 | gpt-5-nano | 2.33 | official | |
| 88 | mistral-nemo | 2.33 | official | |
| 91 | minimax-text-01 | 2.38 | official | |
| 92 | gemini-2.5-flash | 2.39 | official | |
| 93 | claude-opus-4.8 | 2.4 | official | |
| 94 | glm-4.7 | 2.42 | official | |
| 95 | gemini-3-pro-preview | 2.45 | official | |
| 95 | qwen3.7-max | 2.45 | official | |
| 97 | glm-5.2 | 2.48 | official | |
| 98 | minimax-m3 | 2.52 | official | |
| 99 | mimo-v2.5 | 2.525 | self run | |
| 100 | deepseek-v3 | 2.54 | official | |
| 101 | gpt-oss-120b | 2.57 | official | |
| 102 | gemini-3.5-flash-lite | 2.58 | official | |
| 102 | gemini-3.8-flash | 2.58 | official | |
| 104 | gpt-4o-mini | 2.6 | official | |
| 105 | grok-4.6 | 2.62 | official | |
| 105 | mistral-medium-3 | 2.62 | official | |
| 107 | gpt-5-mini | 2.65 | official | |
| 107 | llama-4-maverick | 2.65 | official | |
| 107 | nova-pro | 2.65 | official | |
| 110 | grok-4.1-fast | 2.72 | official | |
| 111 | claude-opus-4.5 | 2.75 | official | |
| 112 | gemini-3.7-flash | 2.77 | official | |
| 113 | glm-4.7-flash | 2.85 | official | |
| 114 | grok-3 | 2.865 | official | |
| 115 | gemma-4-31b-it | 2.88 | official | |
| 116 | gpt-5.3-chat | 3 | official | |
| 117 | gemini-2.0-flash-lite | 3.02 | official | |
| 118 | gemini-3.1-pro-preview | 3.03 | official | |
| 119 | gemini-3.6-flash | 3.05 | official | |
| 120 | gemini-3.1-flash-lite | 3.225 | official | |
| 121 | grok-4.5 | 3.52 | official | |
| 122 | gemini-3.5-flash | 3.88 | official | |
| 123 | gemini-3-flash-preview | 3.9 | official |