{"schema":"postcutoff/event@1","as_of":"2026-10-20T02:15:00+02:00","url":"https://postcutoff.com/e/2026-10-15-arc-agi-3-gemini-4-argon-97-5/","md":"https://postcutoff.com/e/2026-10-15-arc-agi-3-gemini-4-argon-97-5/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-10-15-arc-agi-3-gemini-4-argon-97-5","date":"2026-10-15","date_precision":"day","short_title":"ARC Prize verifies Gemini 4 Argon at 97.5% on ARC-AGI-3 with Google's harness (49.8% standard), the second lab above 95%","deck":null,"takeaway":"On Oct 15, 2026 ARC Prize published verified scores for Gemini 4 Argon: 97.5% on ARC-AGI-3 with the provider-adapter harness ($4.8K) and 49.8% with the standard harness ($5.8K), plus 91.3% on ARC-AGI-2 and 98.5% on ARC-AGI-1.","category":"benchmark","category_label":"Benchmarks","importance":4,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":[{"n":1,"title":"ARC Prize on X: Gemini 4 Argon on ARC-AGI (Verified), Oct 15","url":"https://x.com/arcprize/status/2110768288523251762","type":"official","group":"primary","domain":"x.com"},{"n":2,"title":"ARC Prize on X: Argon standard-harness API issues","url":"https://x.com/arcprize/status/2110768292101042240","type":"official","group":"primary","domain":"x.com"},{"n":3,"title":"ARC Prize: full Gemini 4 Argon results","url":"https://arcprize.org/results/google-gemini-4-argon","type":"official","group":"primary","domain":"arcprize.org"},{"n":4,"title":"ARC Prize on X: Haiku 5.5 on ARC-AGI (Verified), Oct 14","url":"https://x.com/arcprize/status/2110417867288039864","type":"official","group":"primary","domain":"x.com"},{"n":5,"title":"ARC Prize on X: DeepSeek V4.1 Flash on ARC-AGI-3, Oct 13","url":"https://x.com/arcprize/status/2110060578530001231","type":"official","group":"primary","domain":"x.com"},{"n":6,"title":"ARC Prize on X: Reflection Beam 1.0 on ARC-AGI (Verified), Oct 19","url":"https://x.com/arcprize/status/2112230540283658499","type":"official","group":"primary","domain":"x.com"},{"n":7,"title":"ARC Prize on X: Tufa Labs ARC-AGI-3 high score 68.91% (Oct 13)","url":"https://x.com/arcprize/status/2110031207601979715","type":"official","group":"primary","domain":"x.com"},{"n":8,"title":"ARC Prize on X: Tufa Labs ARC-AGI-2 high score 88.06% (Oct 9)","url":"https://x.com/arcprize/status/2108578092558250148","type":"official","group":"primary","domain":"x.com"},{"n":9,"title":"ARC Prize leaderboard","url":"https://arcprize.org/leaderboard","type":"official","group":"primary","domain":"arcprize.org"}],"official":9,"filed":"2026-10-19","updated":"2026-10-19","orgs":["ARC Prize Foundation","Google DeepMind"],"title":"ARC Prize verifies Gemini 4 Argon at 97.5% on ARC-AGI-3 with Google's harness (49.8% standard), the second lab above 95%","summary":"On Oct 15, 2026 ARC Prize published verified scores for Gemini 4 Argon: 97.5% on ARC-AGI-3 with the provider-adapter harness ($4.8K) and 49.8% with the standard harness ($5.8K), plus 91.3% on ARC-AGI-2 and 98.5% on ARC-AGI-1. Google DeepMind is the second lab, after OpenAI's GPT-6 Astra (99.9% / 62.7%), above 95% on ARC-AGI-3. The same week ARC Prize scored Haiku 5.5, DeepSeek V4.1 Flash and Reflection's Beam 1.0, and Tufa Labs reached 68.91% on the compute-limited Kaggle ARC-AGI-3 track.","key_facts":["Gemini 4 Argon (ARC Prize, Oct 15): ARC-AGI-3 97.5% at $4.8K (provider-adapter harness, which preserves opaque reasoning and enables automatic compaction) and 49.8% at $5.8K (standard harness, low reasoning); ARC-AGI-2 91.3% at $0.32/task; ARC-AGI-1 98.5% at $0.16/task (~200k views on X)","Caveat: API issues stopped complete standard-harness testing: low reasoning 49.8% (runs completed on 42 of 55 games), medium 46.3% (46 of 55), high only 19 of 55 so no score; scores count all 55 games with partial credit; costs use Argon's introductory $2/$10 per 1M tokens","Comparison: GPT-6 Astra 99.9% (provider) / 62.7% (standard) on ARC-AGI-3 (Sept 3)","Claude Haiku 5.5 (Oct 14): ARC-AGI-3 2.8% ($437, provider) and 0.5% ($388, standard); ARC-AGI-2 61.3% at $0.085/task; ARC-AGI-1 89.0% at $0.051/task; max reasoning omitted on ARC-AGI-2 after completing only 74 of 120 tasks","DeepSeek V4.1 Flash (Oct 13): ARC-AGI-3 3.3% ($878, provider) and 0.9% ($1.2K, standard), the first model ARC Prize evaluated to exceed 3% on ARC-AGI-3 for under $1K total","Reflection AI Beam 1.0 (Oct 19): ARC-AGI-2 23.5% at $0.61/task, ARC-AGI-1 74.0% at $0.31/task (xhigh reasoning; Baseten pricing $1.30/$4.30 per 1M); ARC-AGI-3 pending","ARC Prize 2026 Kaggle (compute-limited): Tufa Labs set ARC-AGI-2 high score 88.06% (Oct 9; teams above 85% split a $150K bonus) and ARC-AGI-3 high score 68.91% (Oct 13, up from 56%)"],"key_numbers":[],"tags":["arc-agi","arc-agi-3","benchmarks","gemini","agents","reasoning"],"science":null,"body_md":"## What happened\n\nARC Prize posted its verified evaluation of Gemini 4 Argon on Oct 15. With Google's own agent harness, Argon nearly matched GPT-6 Astra on\nARC-AGI-3, the interactive game benchmark launched in March 2026. With the neutral standard harness it reached about half, below Astra's\n62.7%, though API failures left part of the standard runs unfinished. ARC Prize posted further verified results through the week. All\nnumbers here come from ARC Prize's own posts (read via fxtwitter); the result pages were not opened.\n\n## Why it matters\n\nTwo frontier labs now score above 95% on ARC-AGI-3 when they control the harness, while open, compute-limited Kaggle entries passed\ntwo-thirds. The gap between provider and standard harnesses (97.5% vs 49.8%) shows how much of agent performance on the benchmark\ndepends on scaffolding rather than the model alone.","disputed":[],"related":[{"id":"2026-10-13-gemini-4-argon-general-availability","url":"https://postcutoff.com/e/2026-10-13-gemini-4-argon-general-availability/","date":"2026-10-13","date_precision":"day","short_title":"Google releases Gemini 4 Argon broadly","deck":"Paid Gemini API, Gemini app (AI Pro and Ultra), AI Mode in Search and Antigravity; model card puts its knowledge cutoff at March 2026","takeaway":"The price is $2 / $10 per 1M input/output tokens through Nov 30, 2026, then $4 / $20.","category":"model-release","category_label":"Model releases","importance":5,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":56,"official":32,"filed":"2026-10-13","updated":"2026-10-14","orgs":["Google DeepMind","Google"]},{"id":"2026-10-07-claude-haiku-5-5","url":"https://postcutoff.com/e/2026-10-07-claude-haiku-5-5/","date":"2026-10-07","date_precision":"day","short_title":"Claude Haiku 5.5 released at $0.10/$0.50, matching GPT-6 Luna","deck":null,"takeaway":"It ends the gap in Anthropic's lineup at the cheap end, where OpenAI's GPT-6 Luna and Google's Flash-Lite models had undercut Haiku 4.5 by an order of magnitude.","category":"model-release","category_label":"Model releases","importance":4,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":13,"official":8,"filed":"2026-10-08","updated":"2026-10-12","orgs":["Anthropic"]},{"id":"2026-10-05-reflection-beam-501b-open-weight","url":"https://postcutoff.com/e/2026-10-05-reflection-beam-501b-open-weight/","date":"2026-10-05","date_precision":"day","short_title":"Reflection AI unveils Beam, a 501B-parameter Apache-2.0 open-weight MoE","deck":"Weights released Oct 16","takeaway":"This is the first model from the best-funded US lab dedicated to open weights (about $4.6B raised, $25B valuation, more than $7B in GPU deals).","category":"model-release","category_label":"Model releases","importance":4,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":13,"official":5,"filed":"2026-10-05","updated":"2026-10-17","orgs":["Reflection AI"]},{"id":"2026-10-01-arc-prize-2026-milestone-2-winners","url":"https://postcutoff.com/e/2026-10-01-arc-prize-2026-milestone-2-winners/","date":"2026-10-01","date_precision":"day","short_title":"Open-source ARC-AGI-3 agents top out at 27.9% at ARC Prize 2026 Milestone #2","deck":null,"takeaway":"The Kaggle track runs open models under Kaggle's compute limits, so it measures how far open methods get without frontier APIs.","category":"benchmark","category_label":"Benchmarks","importance":2,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":8,"official":8,"filed":"2026-10-04","updated":"2026-10-04","orgs":["ARC Prize Foundation","Tufa Labs"]},{"id":"2026-09-30-gemini-4-argon","url":"https://postcutoff.com/e/2026-09-30-gemini-4-argon/","date":"2026-09-30","date_precision":"day","short_title":"Google announces Gemini 4 Argon, its new frontier model, first released only to cyber defenders via the Fairwind Program","deck":null,"takeaway":"On Sept 30, 2026 Google DeepMind announced Gemini 4 Argon, its first new flagship since Gemini 3.1 Pro.","category":"model-release","category_label":"Model releases","importance":5,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":39,"official":13,"filed":"2026-09-30","updated":"2026-10-13","orgs":["Google DeepMind","Google"]},{"id":"2026-09-03-arc-agi-3-gpt-6-astra","url":"https://postcutoff.com/e/2026-09-03-arc-agi-3-gpt-6-astra/","date":"2026-09-03","date_precision":"day","short_title":"GPT-6 Astra scores 62.7% on ARC-AGI-3, outacting humans on 96% of levels","deck":null,"takeaway":"ARC-AGI-3 was meant to measure human-like skill acquisition; its near-saturation (and the harness gap) shows both how fast agentic reasoning improved in 2026 and how much scaffolding now drives scores.","category":"benchmark","category_label":"Benchmarks","importance":5,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":7,"official":4,"filed":"2026-09-29","updated":"2026-10-04","orgs":["ARC Prize Foundation","OpenAI"]}],"people":[],"posts":[{"id":"x-arcprize-2112230540283658499","title":"“Beam 1.0 from @reflection_ai on ARC-AGI (Verified): - ARC-AGI-2: 23.5%, $0.61/task…”","url":"https://postcutoff.com/p/x-arcprize-2112230540283658499/"},{"id":"x-arcprize-2110768288523251762","title":"“Gemini 4 Argon from @Google on ARC-AGI (Verified): - ARC-AGI-3: 97.5%, $4.8K (provider…”","url":"https://postcutoff.com/p/x-arcprize-2110768288523251762/"},{"id":"x-arcprize-2110768292101042240","title":"“In addition to its strong performance on ARC-AGI-3 using the provider adapter harness…”","url":"https://postcutoff.com/p/x-arcprize-2110768292101042240/"},{"id":"x-arcprize-2110417867288039864","title":"“Haiku 5.5 from @AnthropicAI on ARC-AGI (Verified): - ARC-AGI-3: 2.8%, $437 (provider…”","url":"https://postcutoff.com/p/x-arcprize-2110417867288039864/"},{"id":"x-arcprize-2110031207601979715","title":"“New ARC Prize 2026 - ARC-AGI-3 High Score 68.91% by @tufalabs Tufa Labs jumped from 56% >…”","url":"https://postcutoff.com/p/x-arcprize-2110031207601979715/"},{"id":"x-arcprize-2110060578530001231","title":"“DeepSeek V4.1 Flash from @deepseek_ai on ARC-AGI (Verified): - ARC-AGI-3: 3.3%, $878…”","url":"https://postcutoff.com/p/x-arcprize-2110060578530001231/"},{"id":"x-arcprize-2108578092558250148","title":"“New ARC Prize 2026 - ARC-AGI-2 High Score 88.06% by @tufalabs The $150K Bonus Prize, on…”","url":"https://postcutoff.com/p/x-arcprize-2108578092558250148/"}],"videos":[],"models":[],"changes":[{"date":"2026-10-19","type":"filed","text":"Created. Late; Haiku 5.5, DeepSeek V4.1 Flash, Beam 1.0 and Kaggle high scores added"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-10-19","run":null,"sources_read":null,"updated":"2026-10-19","human_review":null,"version":{"date":"2026-10-19"}},"gaps":[{"model_id":"gpt-6-astra","name":"GPT-6 Astra","cutoff":"2026-04","days_after":168,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":107,"in_training_data":false},{"model_id":"gemini-4-argon","name":"Gemini 4 Argon","cutoff":"2026-03","days_after":198,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":137,"in_training_data":false}],"short_url":"https://postcutoff.com/s/arc-prize-verifies-gemini"}