←Search results

Search “Reasoning”

By relevance · Total 3281 results Clear search

Friday

  2. The Guardian Technology

    Meta releases open-weight model Muse Spark Open

    Meta released the open-weight model Muse Spark Open on Thursday, saying its capabilities can match and in some scenarios surpass comparable products from Chinese vendors. Meta AI head Alexandr Wang called it the most capable U.S. open-weight model to date; the model targets coding and agentic tasks, excels at long-horizon tasks, complex software engineering and multi-step reasoning, and has built-in safety safeguards. The model is optimized for the Apple M5 Ultra chip, and users can run it on the desktop with two Mac Studios equipped with that chip.

    Meta released the open-weight model Muse Spark Open on Thursday, saying its capabilities can match and in some scenarios surpass comparable products from Chinese vendors. Meta AI head Alexandr Wang called it the most capable U.S. open-weight model to date; the model targets coding and agentic tasks, excels at long-horizon tasks, complex software engineering and multi-step reasoning, and has built-in safety safeguards. The model is optimized for the Apple M5 Ultra chip, and users can run it on th

    Matched fieldsSummary、Body

    #hot-signal#Model releases#Open-source ecosystem#Meta
  3. Hugging FaceSelected

    Hugging Face Releases Beam: A 501B Open Source Model Supporting 1M Context and Self-Hosting

    Hugging Face has released Beam, a 501B-parameter sparse MoE open source model aimed at long-text reasoning and agent tasks with 1M context, activating only about 23B parameters per token. The model uses 52 layers of hybrid attention, with 43 layers using a 512-token sliding window and 9 layers using global attention; the KV cache grows by 36 KiB per token, about one-sixth that of a full-attention approach. The official release provides three checkpoints—BF16, FP8, and NVFP4; NVFP4 can compress the weights from about 1 TB to about 280 GB, and day-0 support is already available in transformers, vLLM, and SGLang.

    Hugging Face has released Beam, a 501B-parameter sparse MoE open source model aimed at long-text reasoning and agent tasks with 1M context, activating only about 23B parameters per token. The model uses 52 layers of hybrid attention, with 43 layers using a 512-token sliding window and 9 layers using global attention; the KV cache grows by 36 KiB per token, about one-sixth that of a full-attention approach. The official release provides three checkpoints—BF16, FP8, and NVFP4; NVFP4 can compress t

    Matched fieldsSummary、Body

    #lab#Industry news#Open-source ecosystem#Hugging Face
  4. Ollama BlogSelected

    Reflection's Open-Weight Model Beam Launches on Ollama, Free for a Limited Time

    Reflection's first open-weight model, Beam, is officially released, offered free for a limited time with Ollama as a day-zero partner and with zero data retention (ZDR). Beam is an MoE model with approximately 501 billion total parameters, about 23 billion activated per token, and a 1 million token context window. It targets software engineering, reasoning, and long-horizon agentic tasks, and is released under the Apache 2.0 license. Users can run beam:cloud via Ollama Cloud and connect it to tools such as Claude Code, Codex, and OpenCode.

    Beam agentic coding benchmarks

    Reflection's first open-weight model, Beam, is officially released, offered free for a limited time with Ollama as a day-zero partner and with zero data retention (ZDR). Beam is an MoE model with approximately 501 billion total parameters, about 23 billion activated per token, and a 1 million token context window. It targets software engineering, reasoning, and long-horizon agentic tasks, and is released under the Apache 2.0 license. Users can run beam:cloud via Ollama Cloud and connect it to to

    Matched fieldsSummary、Body

    #company#Product updates#Open-source ecosystem#English text is being prepared.
  5. Baseten Engineering BlogSelected

    Reflection's Open-Weight Model Beam Launches on Baseten: 501B MoE with 1M Context

    Reflection's first open-weight model, Beam, is now available on the Baseten platform, with day-0 support through Baseten Model APIs and Dedicated Inference. Beam uses a MoE architecture with 501B total parameters and 23B activated per token, a context window of up to 1 million tokens, and is aimed at coding and agent workflows. During training, Reflection used longer RL rollouts and added a length penalty to encourage the model to complete tasks with fewer tokens; users can also adjust reasoning effort via a parameter to control how much reasoning the model invests.

    Beam

    Reflection's first open-weight model, Beam, is now available on the Baseten platform, with day-0 support through Baseten Model APIs and Dedicated Inference. Beam uses a MoE architecture with 501B total parameters and 23B activated per token, a context window of up to 1 million tokens, and is aimed at coding and agent workflows. During training, Reflection used longer RL rollouts and added a length penalty to encourage the model to complete tasks with fewer tokens; users can also adjust reasoning

    Matched fieldsSummary、Body

    #editorial#Product updates#Reasoning#English text is being prepared.
  6. Artificial Analysis full articleSelected

    Reflection Releases Its First Model Beam, Scoring 30 on the Artificial Analysis Intelligence Index

    Reflection has released its first model, Beam, an open-weight model with 501B total parameters and 23B activated parameters. It scores 30 on the Artificial Analysis Intelligence Index, ranking second among US open-weight models, behind only Meta's Muse Spark Open (39) and ahead of NVIDIA's Nemotron 3.5 Super VL (26). Beam consumes about 26k output tokens per Intelligence Index task, of which about 19k are used for reasoning and 7k for answering; it scores 48.5% on AutomationBench-AA, but only 3.0% on Terminal-Bench 4.0. On AA-Omniscience, it answers 33% of questions correctly, with a hallucination rate of 47% and a net index score of 1.5.

    Reflection has released its first model, Beam, an open-weight model with 501B total parameters and 23B activated parameters. It scores 30 on the Artificial Analysis Intelligence Index, ranking second among US open-weight models, behind only Meta's Muse Spark Open (39) and ahead of NVIDIA's Nemotron 3.5 Super VL (26). Beam consumes about 26k output tokens per Intelligence Index task, of which about 19k are used for reasoning and 7k for answering; it scores 48.5% on AutomationBench-AA, but only 3.

    Matched fieldsSummary、Body

    #company#Model releases#Open-source ecosystem#Evaluations / benchmarks
  12. 36Kr · AISelected

    DeepMind AlphaProof Nexus appears in《Science》: autonomously solves 9 Erdős problems

    Google DeepMind's automated mathematical proof system AlphaProof Nexus appeared in the October 8 issue of《Science》, autonomously solving 9 of 353 open Erdős problems, two of which had remained open since 1970. The system uses Gemini 3.1 Pro to draft proofs in Lean, with a compiler verifying line by line and feeding back iterations; the reasoning cost per problem is several hundred dollars. The paper's ablation results show that a simplified system that removes complex structures such as the sub-agent tournament and retains only the large model plus Lean feedback can also solve these 9 problems, though the hardest few are more expensive.

    Google DeepMind's automated mathematical proof system AlphaProof Nexus appeared in the October 8 issue of《Science》, autonomously solving 9 of 353 open Erdős problems, two of which had remained open since 1970. The system uses Gemini 3.1 Pro to draft proofs in Lean, with a compiler verifying line by line and feeding back iterations; the reasoning cost per problem is several hundred dollars. The paper's ablation results show that a simplified system that removes complex structures such as the sub-

    Matched fieldsSummary、Body

    #hot-signal#Papers/Research#Reasoning#DeepMind

Thursday

  4. OpenRouter

    NVIDIA Nemotron 3.5 Super VL Launches on OpenRouter

    NVIDIA Nemotron 3.5 Super VL is now available on OpenRouter. It is an open-source 120B-A12B multimodal model designed for long-running agents that need to observe, reason, and act across text, images, documents, GUIs, and video, with support for a 1M context. The model uses a Hybrid Mamba-Transformer MoE architecture and supports multi-token prediction to speed up output, targeting scenarios such as document intelligence, web research, customer service, and video understanding.

    Matched fieldsSummary

    #hot-signal#Product updates#Multimodal#Agents