Reactions from ranked influencers
6 posts
Inkling from @thinkymachines on ARC-AGI (Verified) - ARC-AGI-2: 36.5%, $0.64/task - ARC-AGI-1: 79.5%, $0.30/task As of today, Inkling is the highest-scoring open-weight model evaluated by ARC Prize on both ARC-AGI-1 and ARC-AGI-2.
Inkling’s humility at launch may have made it underrated. The previewed small model will wake a lot of people up (if it comes soon). Look at its peers on ARC AGI.
Inkling from @thinkymachines on ARC-AGI (Verified) - ARC-AGI-2: 36.5%, $0.64/task - ARC-AGI-1: 79.5%, $0.30/task As of today, Inkling is the highest-scoring open-weight model evaluated by ARC Prize on both ARC-AGI-1 and ARC-AGI-2.
Inkling is the smartest open-weight model right now but it only has ~1-2 weeks until Kimi-K3 weights are out
Inkling from @thinkymachines on ARC-AGI (Verified) - ARC-AGI-2: 36.5%, $0.64/task - ARC-AGI-1: 79.5%, $0.30/task As of today, Inkling is the highest-scoring open-weight model evaluated by ARC Prize on both ARC-AGI-1 and ARC-AGI-2.
Inkling’s humility at launch may have made it underrated. The previewed small model will wake a lot of people up (if it comes soon). Look at its peers on ARC AGI.
Inkling from @thinkymachines on ARC-AGI (Verified) - ARC-AGI-2: 36.5%, $0.64/task - ARC-AGI-1: 79.5%, $0.30/task As of today, Inkling is the highest-scoring open-weight model evaluated by ARC Prize on both ARC-AGI-1 and ARC-AGI-2.
We are excited that Inkling is the highest-scoring open-weight (as of today) model evaluated by @arcprize on both ARC-AGI-1 and ARC-AGI-2!
Inkling from @thinkymachines on ARC-AGI (Verified) - ARC-AGI-2: 36.5%, $0.64/task - ARC-AGI-1: 79.5%, $0.30/task As of today, Inkling is the highest-scoring open-weight model evaluated by ARC Prize on both ARC-AGI-1 and ARC-AGI-2.
So we know that a 1e25ish pretrain with mediocre post-training, the scale of V4 or others in this bracket, is enough for 36.5% on ARC-AGI-2. GLM-5.2 did 22.8%. Grok's crappy 500b can do >60%. I'm not sure this says much
Inkling from @thinkymachines on ARC-AGI (Verified) - ARC-AGI-2: 36.5%, $0.64/task - ARC-AGI-1: 79.5%, $0.30/task As of today, Inkling is the highest-scoring open-weight model evaluated by ARC Prize on both ARC-AGI-1 and ARC-AGI-2.