The Open Weights
LatestModelsLeaderboardsCompanies
Subscribe
The Open Weights

The daily record of open-source AI. New model releases, leaderboards, and what's coming next — written for people who ship.

Refreshed every 12 hours

Discover

  • Latest releases
  • New today
  • Trending models

Browse

  • All models
  • Companies
  • Categories
  • Leaderboards

About

  • About
  • Editorial policy
  • Hardware estimates
  • RSS feed
  • llms.txt
  • Newsletter

© 2026 The Open Weights. An independent publication.

PrivacyTermsSMSAggregated by Claude · curated by humans.
LatestReflection AI1
Reflection AIText / LLM

Reflection AI debuts Beam, a 501B open model

The startup's first open-weight release is a dense 501-billion-parameter model aimed at text and reasoning tasks.

Oct 5, 2026
NotableOther

Reflection AI has released Beam, its first open-weight language model, according to the company's announcement. The model carries 501 billion parameters and is positioned for both general text generation and reasoning workloads.

Notably, Beam is a dense model rather than a mixture-of-experts design — a choice that sets it apart from many recent frontier-scale releases, which have leaned on sparse architectures to keep inference costs down. A dense model at this size signals a bet on raw capability, though it also implies substantial hardware requirements to run.

Why it matters

Open-weight models in the half-trillion-parameter range remain rare, and each new entrant expands the options available to researchers and builders who want to inspect, fine-tune, or self-host large systems rather than rely on closed APIs.

  • Scale: 501B parameters, dense architecture
  • Focus: text generation and reasoning
  • Access: open weights under a custom license

Key details — including context length and benchmark results — were not specified in the release record, and the model ships under a custom license rather than a standard open-source one, so teams will want to review the terms before building on it. Reflection's blog post is the authoritative source for the full picture.

Sources

  • Beam: Reflection's 501B open-weight model

    Hacker News

    Visit
OlderNVIDIA's Nemotron 3.5 Bolt Targets 4GB DevicesNVIDIA · Text / LLM · 2 months agoNewerFalcon-Emirati tunes an LLM for local dialectNVIDIA · Text / LLM · last month

Get the model

Hacker News

Specs

Parameters501B
LicenseOTHER

Can you run it?

Needs a multi-GPU server — about 307.6 GB at 4-bit.

  • BF16 (as published)

    Multi-node cluster

    1.0 TB
  • 8-bit

    8×H100 80 GB node

    538.1 GB
  • 4-bit

    4×H100 80 GB · 512 GB Mac Studio

    307.6 GB
Your machine
  • Precision not published; assuming 16-bit (BF16) weights.
  • Parameter count is an editorial estimate — no weights metadata on Hugging Face.
  • KV cache approximated from model size (no config metadata) at 8,192 tokens.

Estimated from the parameter count. Assumes an 8K context and ~1 GB runtime overhead; actual needs vary. How we estimate


Modalities

Text / LLMReasoning
2 versions — view changelog

The Weekly Weights

Every open release that mattered, one email a week.

0 comments

No comments yet. Be the first to weigh in.

More in Text / LLM

All Text / LLM →
Aion-1.0-Instruct-2.3B
Microsoft/Text / LLMRuns anywhere

Microsoft's Aion 1.0 Instruct is a compact 2.3B chat model

A small, multilingual instruction-tuned model aimed at chat and summarization lands on Hugging Face.

Nov 18, 2026
SenseNova-Loop-9B
SenseTime/Vision-LanguageRuns on a laptop

SenseTime Releases Looped 9B Multimodal Model

The company's new SenseNova-Loop-9B pairs vision-language capabilities with a recurrent looped-transformer design.

Nov 17, 2026
Hunyuan Hy4-flash
Tencent/Vision-LanguageMulti-GPU server

Tencent releases Hunyuan Hy4-flash multimodal MoE

The lightweight, mixture-of-experts model adds vision and text instruction-following to Tencent's open Hunyuan lineup.

Nov 17, 2026