#Reasoning

AIOpenAI published 722 math manuscripts covering 372 result groups in a new GitHub repository, openai/math, all produced by an unreleased internal model. The author describes the results as including a near-Riemann hypothesis claim pushed to 0.875, and notes that 25 Fields Medal winners criticized the company's approach to AI math research.

Why it matters: The piece traces how AI math results moved from benchmarks to open problems, offering context on verification and the mathematicians' pushback.

AIEpoch gave six models 11 real work tasks from its own operations, including graphic design, data insights, and research design, and graded outputs against employee standards. Fable 5.1 and GPT-6 Astra led on average task performance, reliably handling well-defined work such as coding and computational analysis. The report finds that all models still fail on open-ended judgment, including matching Epoch's standards, designing informative experiments, and generating diverse ideas, so the authors conclude AI cannot yet replace workers at Epoch.

Why it matters: The report separates well-defined task reliability from open-ended judgment failures, which benchmark scores on easily verifiable tasks would miss.

AINVIDIA reports that fine-tuned Nemotron models reached gold-medal level at both IOI 2026, scoring 535.4 out of 600, and IMO 2026, scoring 30 out of 42. The IOI run was a live, unofficial, unsupervised benchmark, while IMO proofs were graded by official IMO graders. The post also releases checkpoints, datasets, a new 200-problem benchmark, and inference pipelines on Hugging Face and NeMo-Skills.

Why it matters: The post traces how SFT, RL, and a generate-verify-refine loop turned Nemotron into gold-level specialists for IOI and IMO, with the training and inference details shared.

AIEpoch AI's InnovationEval tested whether AI agents could independently devise a post-training method matching on-policy self-distillation (SDPO), a recent human-developed innovation. GPT-5.6 Sol achieved only a small in-scope gain, about 15% of SDPO's gains after adjustment, and Claude Fable 5 mainly reported gains from selecting the best of several runs, which were excluded as out of scope. The authors conclude that current models have not yet independently discovered a meaningful AI algorithmic innovation.

Why it matters: The evaluation tests whether AI can independently devise a post-training method matching a published human innovation, with a scope and memorization caveat worth reading.

AIReflection AI introduced Beam, an agentic open model with 501B total parameters and 23B active parameters, trained end-to-end from scratch. The quoted announcement says it targets frontier reasoning efficiency and coding and agentic tasks, with full weights due this month. Clément Delangue, Hugging Face's CEO, reposted it with a welcome to the Reflection organization on Hugging Face.

Why it matters: The quoted announcement names Beam's parameter scale, active-parameter count, and coding and agentic focus, which helps readers gauge where it fits among open models.

AIHugging Face and collaborators published a guide to multi-harness RL that trains models through a capture proxy without changing the agent harness. The proxy records the token ids and logprobs vLLM samples, and the source reports LFM2.5-2.6B rising from 42% to 54% after training across four harnesses. Fine-tuning on 3,189 successful rollouts from Qwen3.8-27B plateaued at 47.5%, below both RL runs, and the capture proxy, trainer, tasks, SFT data, training code, and seven trained models are released openly.

Why it matters: The source gives a concrete method for training models across several agent harnesses, with measured gains and a note that imitation learning underperformed RL.

AIPhysicist Matthew Schwartz describes building BootLoops, an open-source harness for exact quantitative calculations, after choosing problems suited to Claude's strengths. He reports that Claude solved long-standing integrals and found connections across ecology, population genetics, economics, and linguistics, with domain experts steering results toward questions those fields care about. The post states that the approach required constant human oversight, since Claude often overstated results and misjudged time.

Why it matters: The guest post explains why scientists often find current AI tools frustrating and offers a method for finding problems where AI and researchers match, backed by concrete projects.

AIGoogle AI announced Gemini 4 Argon, a new frontier model built for deep reasoning across long, complex workflows in software engineering, legal and finance knowledge work, and cybersecurity defense. Google says it is expanding the model's output token limit to 1M tokens. Argon is rolling out first to trusted cyber defenders in the Fairwind Program, with broader availability to follow as soon as possible.

Why it matters: The benchmark table compares Gemini 4 Argon against GPT-6 Astra and Claude models across knowledge work, coding, and multimodal tasks, showing where it leads and trails.

AIGoogle DeepMind announced Gemini 4 Argon, rolling out first to trusted cyber defenders through its Fairwind Program. Argon will launch at an introductory price of $2 per million input tokens and $10 per million output tokens, with output limits raised to 1M tokens. The post cites a 77.9% score on DeepSWE v1.1 and 91.7% on LVBench, and says broad availability will follow safeguard testing.

Why it matters: The post pairs Argon's benchmark claims with the phased release, pricing, and safeguard details, helping readers weigh its frontier-level capabilities against its access limits.

AIArtificial Analysis reports that Google's Gemini 4 Argon scores 53 on its Intelligence Index with high reasoning, matching GPT-6 Astra (max) and one point ahead of GPT-6.1 Sol (max). At the current 50% launch discount, its cost per task is $1.99, about 60% of GPT-6 Astra's $3.26, but the discount's end date is unconfirmed and standard pricing would raise it to $3.98. The model is being rolled out to selected users and is not publicly available.

Why it matters: The benchmark compares Gemini 4 Argon's cost per task and hallucination rate with GPT-6 Astra, showing where its value depends on a temporary 50% discount.

AIBAAI released AREX-2, a 27B-parameter long-horizon agent model that improves solutions over multiple test-time rounds by proposing, measuring, reflecting, and revising. It was trained on machine-learning and algorithmic-programming tasks with verifiable feedback, and the source reports that this self-improvement transfers to deep research. The model is Apache License 2.0 licensed and has a 262,144-token context length.

Why it matters: The source compares AREX-2 against closed and open models on coding and deep-research benchmarks, showing how test-time self-improvement is measured across task types.

AIResearchers from Tsinghua NLP and collaborators show that on-policy distillation with a single training query recovers 87% of full-data gains on math, reaching 68.5 versus 69.8 by step 300. The paper attributes the slow progress to how fast the student absorbs the teacher's signal rather than to dataset size. Code and the paper are publicly available on GitHub and Hugging Face.

Why it matters: The paper isolates training data from the algorithm, showing one query nearly matches full-data on-policy distillation, which reframes where post-training gains come from.

AIOpenAI released GPT-6 Sol and Luna, priced 50 percent below GPT-5.6 promo API pricing, and rolling out in ChatGPT Work, Codex and the API, not yet in regular Chat. Anthropic released Claude Opus 5.5, described as roughly Claude Fable 5.1 level for 40 percent less than Opus 5 and over 30 percent faster, with Sonnet 5.5 and Haiku 5.5 due in coming weeks.

Why it matters: The recap puts OpenAI and Anthropic releases side by side, with pricing and capability claims that help compare the two launches.

AIXiaomi MiMo reports that tool-call repetition in MiMo-V2.6 reached over 0.05% of responses across agent harnesses, causing stalled agents and wasted context. The team traced the cause to an RL flooding penalty set at 32 calls per turn, which missed smaller excess behavior, and replaced the approach with a specialized teacher distilled via MOPD. Repetition rates for both Pro and Flash dropped substantially, at roughly $90,000 versus an estimated $2.31 million for the alternative fix.

Why it matters: The post traces an agent failure to a reward blind spot and compares the costs of two fixes, offering a transferable debugging method for RL-trained tool-calling models.

AIAnthropic reports that Claude solved a nine-loop calculation in the planar N=4 super-Yang-Mills model, surpassing the previous eight-loop record set by Lance Dixon and collaborators. The quoted post says Claude ran largely unsupervised for days in Claude Science using a single prompt, at a total cost of a few thousand dollars, and Dixon independently verified the result. Kevin Weil's own text praises the achievement and expects AI to advance high energy physics over the coming 12 months.

Why it matters: The quoted Anthropic post gives a concrete benchmark: Claude ran for days to reach nine loops, extending the previous eight-loop record in a physics model.

AIAnthropic reports that Claude solved a nine-loop scattering amplitude problem in planar N=4 super-Yang-Mills, surpassing the previous eight-loop record set by SLAC's Lance Dixon and collaborators. Working largely unsupervised for days from a single prompt, at a total cost of a few thousand dollars, Claude used methods developed by Dixon's group, and Dixon independently verified the result.

Why it matters: The post shows Claude solving a nine-loop physics calculation beyond the previous eight-loop record, verified independently, which bears on AI use in theoretical physics research.

AIFireworks Research released Ember-1, a specialized model built on Kimi K3 that it says delivers the same quality with 40% fewer tokens. Across five industry benchmarks, Ember-1 matched K3 max quality at a fraction of the cost, and in two customer A/B tests it used about 35% fewer tokens per task. It is available as a Research Preview on Serverless, and Fireworks is also launching training support for customized models.

Why it matters: The source gives benchmark and A/B results for cutting reasoning tokens while holding quality, which bears on cost planning for coding and agent workloads.

AIXiaomi MiMo announced open-source releases of Pro and Flash, the MiMo-V2.6-Distill-Qwen-9B model, a technical report, over 7K RL task environments, an end-to-end RL framework, and composable mini-harnesses. The attached table shows MiMo-V2.6-Distill-Qwen-9B after SFT and after RL compared with Qwen3.5-9B, with RL scores higher on most listed benchmarks, such as SWE-bench Verified at 66.2 versus 60.0.

Why it matters: The table compares a 9B distilled model against Qwen3.5-9B on coding, cyber, and agent benchmarks, showing how the reinforcement learning stage changes results.

AIXiaomi released MiMo-V2.6-Flash-RL, an efficiency-balanced checkpoint in its MiMo-V2.6 series, on Hugging Face. The model is a sparse MoE with 309B total and 15B activated parameters, supports text, image, video, and audio input, and offers a 1M-token context. The technical report says it was trained with a single mixed reinforcement learning run across coding, agent, visual, and cybersecurity tasks.

Why it matters: The report pairs its benchmark tables with the RL training method, which helps readers judge how the checkpoint's scores relate to its training approach.

AIXiaomi MiMo released MiMo-V2.6-Pro-RL on Hugging Face, a sparse MoE model with 1.02T total and 42B activated parameters and a 1M-token context. The technical report says it accepts text, image, video, and audio, and was trained with a single mixed reinforcement learning run across coding, agent, visual, and cybersecurity tasks.

Why it matters: The report pairs a 1.02T-parameter MoE model with an RL-based self-improvement method, useful for judging how reinforcement learning is scaled in frontier open models.