ALoDLM
PulseGate's liveness check found it on 6 Oct 2026; it is registered on GitHub and has been in the index since 6 Oct 2026. How this is checked
ALoDLM is a research project for diffusion language models that use token-adaptive latent recurrence during denoising. It provides 1.7B and 8B model variants, code, evaluation results, and decoding demonstrations for researchers and developers exploring parallel text generation.
Inferred · not functionally tested
Overview
6 featuresPurpose: Reducing the quality and efficiency gap between parallel diffusion language models and autoregressive text generation.
Inferred · not functionally tested
Audience: AI researchers and machine learning developers
Inferred · not functionally tested
Functions: code_generation
Inferred · not functionally tested
Interfaces: API: indicated (inferred, not tested) · MCP: unknown · CLI: indicated (inferred, not tested) · Self-hosting: indicated (inferred, not tested)
Recorded constraints: pricing: open_source · license: Open Source · platforms: WEB · deployment: browser, self_hosted, cli, api_only
Constraint provenance is unknown; confirm requirements with the publisher.
Record sources: alo-dlm.github.io · github.com. These links do not verify the individual claims.
ALoDLM is a Quantised & converted weights project. Inferred · not functionally tested: It focuses on reducing the quality and efficiency gap between parallel diffusion language models and autoregressive text generation. Inferred · not functionally tested: It is built as an open-source project for AI researchers and machine learning developers. Basis unknown · not verified: The project is open source (Open Source). Basis unknown · not verified: ALoDLM is available on the web, the command line, and API, and it can be self-hosted.
It is developed by Amazon AGI (United States), and it first shipped in 2026. The project is developed in the open on GitHub with 12 stars and 5 commits in the last 90 days. Inferred · not functionally tested: Key capabilities include adaptive recurrence, parallel decoding, and 1.7B model.
Summary written by a language model from the project’s public pages.
Tasks: Inferred · not functionally tested
- Adaptive recurrence
- Parallel decoding
- 1.7B model
- 8B model
- Decoding replay
- GSM8K evaluation
Topics: Inferred · not functionally tested
Built with & integrations
- local_oss
- bvllm in the HTML
Trust & compliance
Indexing history
1What PulseGate has recorded for this listing
- Indexed6 Oct · 20:46 UTCALoDLM: Adaptively Looped Diffusion Language Models seen via Hacker News firehose (Algolia)Source: Hacker News firehose (Algolia) · Open
Frequently asked questions about ALoDLM
- What does ALoDLM do?
- Inferred · not functionally tested: ALoDLM focuses on reducing the quality and efficiency gap between parallel diffusion language models and autoregressive text generation. It is catalogued under Quantised & converted weights on PulseGate.
- Who is ALoDLM for?
- Inferred · not functionally tested: ALoDLM is an open-source project built for AI researchers and machine learning developers.
- Does ALoDLM have a free plan?
- Basis unknown · not verified: Yes — ALoDLM is open source under the Open Source license and free to use.
- What platforms does ALoDLM run on?
- Basis unknown · not verified: ALoDLM runs on the web, the command line, and API. It can also be self-hosted.
- Is ALoDLM still maintained?
- PulseGate's liveness check found it on 6 Oct 2026. Its GitHub repository shows 5 commits in the last 90 days.
- What are alternatives to ALoDLM?
- Similar projects tracked by PulseGate include D1 3b, M2 2B, and Qwen3.6 35B A3B Westernized.D1 3bM2 2BQwen3.6 35B A3B Westernized
- Who develops ALoDLM?
- ALoDLM is developed by Amazon AGI, based in the United States.
- How long has ALoDLM been around?
- ALoDLM first shipped in 2026.
Also in Quantised & converted weights
Same category — not a similarity match