←GitHub
← All topics

GitHub

English text is being prepared.

Public reading groups
38
Last 30 days
19
Directory status
Legacy topic

LatestArtificial Analysis tests GitHub Copilot Fusion: $5.59 cost per task…

Latest reading groups

Duplicate reports of the same fact are grouped

  1. Artificial AnalysisAnalysis
    AI score71

    Artificial Analysis tests GitHub Copilot Fusion: $5.59 cost per task…

    In Coding Agent Index v2.0, Artificial Analysis benchmarked GitHub Copilot's Copilot Fusion, which costs $5.59 per task, compared with $11.42 for Claude Code with Opus 5.5 (max), while performing only 2 points lower. Copilot Fusion is a runtime orchestration system that selects and coordinates among Anthropic and OpenAI models such as Claude Opus 5.5, Claude Sonnet 5.5, GPT-6 Astra, and GPT-6 Sol. It leads on Terminal-Bench 4.0 with a score of 71, ranks third on FrontierCode with 51 points, and uses 7.2M tokens per task, about half that of Opus 5.5 in Claude Code.

    Why read thisArtificial Analysis benchmarked Copilot Fusion in the Coding Agent Index, reporting cost, token usage, and scores on three benchmarks, directly comparable with Claude Code and Devin Fusion.

  1. Unite.AIResearch
    AI score85

    Axiom Math releases a 43.1-million-line Lean formal proof of the Classification of Finite Simple Groups

    On October 27, 2026, Axiom Math released what it calls the first complete computer-verified proof of the Classification of Finite Simple Groups (CFSG), written by dozens of AI agents directed by engineers and mathematicians, who produced approximately 43.1 million lines of Lean code in 24 days. The project page lists 2.05 million declarations and a corpus of 715 papers and books; the code is a sorry-free Lean4 proof, pinned to version 4.35.0-rc3, built on Mathlib and TauCeti, and licensed under Apache-2.0. Axiom says that during the formalization it found 34 notable deviations in the public proof, the most substantive of which involves the Unbalanced Group Theorem, and says that AxiomProver produced a new theorem, Theorem 2.3, to patch it.

    Why read thisThe original article records that a cluster of AI agents completed a 43.1-million-line Lean formal proof within 24 days, and discloses 34 deviations in the public proof.

  1. Electronic Technology Application
    AI score94

    OpenAI announces three breakthroughs by its model in mathematics, computer science, and physics

    OpenAI announced that its AI model has achieved three breakthroughs in the fields of mathematics, computer science, and physics, claiming that the model can independently discover and prove complex scientific theories. The mathematics and computer science results have been formally verified through Lean, and the proof materials and natural-language proofs have been uploaded to a public GitHub repository; the physics result is being reviewed by experts including Nobel laureates in physics. In mathematics, the model provided a proof of the local smoothing conjecture in all dimensions, and thereby solved the Kakeya conjecture and the Stein restriction conjecture in all dimensions; in computer science, the model provided a proof that the Exponential Time Hypothesis (ETH) is false. OpenAI stated that the relevant results still require further independent confirmation by the academic community.

    Why read thisThe original text lists the verification methods for the three results and the list of external experts, so readers can use them to judge the credible boundaries of these claims.

  1. 36Kr · AIProduct release
    AI score82

    DeepSeek open-sources inference framework DeepInfer and Token codec component deepseek-recipe

    DeepSeek open-sources its self-developed inference framework DeepInfer (Apache 2.0) and the production-grade Token codec component deepseek-recipe (MIT), supporting DeepSeek V4.1 Flash in serving 155 trillion Tokens per day on a single cluster. Official online statistics show an average of 77K Tokens input per request, a time to first token of about 400 milliseconds, 230 Tokens output per second during the generation phase, and an average of 96% of input hitting the cache. DeepInfer currently supports V4.1 Flash; the open-source version provides only offline inference, and the KV cache management and load balancing code has not yet been fully open-sourced because it is tied to internal infrastructure.

    Why read thisThe original text discloses details of DeepInfer's architecture split, cache hits, and expert parallelism, which can be used to understand the specific practices for reducing costs in large-scale inference.

  1. figma_ai_agent_makeFirst-partyProduct release
    AI score62

    English translation is being prepared.

    A Pickup App preview on the Figma canvas shows a wall clock listing alongside its App.tsx source code.