Exciting news: GPT-6.1 Sol (Max) by @OpenAi just landed in the Agent Arena at #5 (+11.23%) and reshaped the Pareto frontier! At a $0.56 median cost per task, it delivers performance within 2 percentage points of GPT-6 Sol and GPT-6 Astra for substantially less cost: - 39% lower cost than GPT-6 Sol, while scoring +1.52 pts higher - 81% lower cost than GPT-6 Astra, while landing within 1.04 pts GPT-6.1 Sol also delivers top-five performance at substantially lower cost compared to: - 88% lower cost than Claude Fable 5.1 (Max), while landing within 3.08 pts (ranked #1) - 65% lower cost than Claude Opus 5.5 (High), while landing within 2.59 pts (ranked #2) - 80% lower cost than Claude Sonnet 5.5 (Max), while landing within 1.29 pts (ranked #3) Congrats to the team @OpenAI on this release!

Claude Sonnet 5.5 by @AnthropicAI just landed at #3 in the Agent Arena. This model has a median cost per task of $2.74, and a +12.5% net improvement score. Claude Sonnet 5.5 delivers top-tier performance, but at a cost premium: #2 Claude Opus 5.5 costs $1.58 per task while achieving a higher score. That tradeoff keeps Sonnet 5.5 just off the Agent Arena Pareto frontier.

Exciting news: Claude Sonnet 5.5 (Max) by @AnthropicAI has debuted at #3 in the Agent Arena with +12.5% net improvement! This release is a 8.1 percentage-point increase over Claude Sonnet 5 (High), which ranks #13 with +4.4% net improvement. By category, Claude Sonnet 5.5 secured the #1 spot in Chat (+15.6%) above both Fable 5.1 (+11.49%) and Opus 5.5 (+10.29%). This performance comes with a higher cost: Claude Sonnet 5.5 (Max) has a median cost of $2.74 per task, about 73% higher than #2 Claude Opus 5.5 (High) at $1.58. @AnthropicAI models now hold all three top positions in Agent Arena. Congrats to the team!

How to design rewards for post-training frontier image models? Our research suggests human preference reward is necessary, but insufficient: A preference model may still reward outputs that look appealing but miss details, introduce unrequested content, or exhibit other forms of reward-hacking. We therefore optimize towards a composite reward: - Bradley-Terry reward model trained on ~5.6M pairwise human votes - Faithfulness reward from auto-generated prompt checklists evaluated by a vision-language model - Constraint reward covering explicit and implicit user intent - Anti-reward-hacking rubric rewards targeting failures such as garbled text and photorealism drift This post-training recipe improves two already-strong open image models: - Post-trained FLUX.2-dev gains 69 Elo points on our live T2I leaderboard, scoring 1202 - Post-trained Ideogram 4 gains 20 Elo points reaching a score of 1224 and surpassing all publicly listed open models (as of Sep 04, 2026). Offline ablations with Gemini 3.5 Flash as the judge, show that these reward components are complementary: win rate against the base model increases as we add faithfulness and then constraint rewards on top of preference-only training, reaching 64.2%. Finally, we ensemble policies trained with and without the anti-reward-hacking objective directly in weight space, further increasing win rate to 66.0%.

MiMo-V2.6-Pro and MiMo-V2.6-Flash by @XiaomiMiMo have landed in Agent Arena! MiMo-V2.6-Pro ranks #5 among open-source models with +3.17% net improvement across 8.1K+ real-world agentic sessions. This release is a +10.4 percentage-point lift and a 9-spot improvement from MiMo-V2.5-Pro at #13 with -7.23%! Its strongest signal is Confirmed Success (explicit user feedback that the task worked) where it scored +7.35% and ranks #2 among open-source models! MiMo-V2.6-Flash landed #9 among open models with -0.57% net improvement across 13K+ real-world agentic sessions. At a $0.04 median cost per task (56% lower than MiMo-V2.6-Pro), it also landed on the Agent Arena Pareto frontier! See its placement below. Congrats to the @XiaomiMiMo team on the release!

Exciting update: Claude Sonnet 5.5 with xHigh reasoning has landed in the Code Arena: WebDev. With 1786 pts, its ranked #3! At a blended $8/M tokens, Claude Sonnet 5.5 remains on the Pareto frontier with xHigh reasoning. This release is just 2 pts from GPT-6 Astra in the #2 spot with 1788 pts, for 80% of the price. By domain, Claude Sonnet 5.5 (xHigh) landed: - #2 in Gaming, Reference-Based Design, and Brand & Marketing - #3 in Simulations - #4 in Content Creation Tools and Consumer Product - #6 in Data & Analytics Congrats again to @AnthropicAI on this release!

Hidream-O1-Video-1.0 by @HiDream_AI just landed in the Image-to-Video Arena at #7 with 1456 pts! This model is available on @vivago_ai, is 8 pts away from gemini-omni-flash and within 20 pts of both dreamina-seedance-2.0 and 2.5. Congrats to @HiDream_AI on this release!