PulseAugur
EN
LIVE 01:34:18

Inception Labs' Mercury 3 diffusion LLM hits 1,000 tokens/sec

By PulseAugur Editorial · [1 sources] ·

Inception Labs has released Mercury 3, a diffusion-based large language model that achieves significantly higher throughput than traditional autoregressive models. This new model can generate approximately 1,068 tokens per second on NVIDIA GPUs, which is reportedly 4 to 8 times faster than optimized models like GPT-6 Luna and Claude Haiku 5.5. Mercury 3 utilizes a parallel denoising process, starting with a masked draft and iteratively refining all parts of the output simultaneously, allowing for more efficient GPU utilization. The model also boasts a 1-million-token context window and is priced competitively for high-throughput applications. AI

IMPACT This diffusion LLM's high throughput could enable new applications and lower costs for high-volume text generation tasks.

RANK_REASON New model release from a lab (Inception Labs) with a novel architecture (diffusion LLM) and performance claims. [lever_c_demoted from frontier_release: ic=1 ai=1.0]

Read on dev.to — LLM tag →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Inception Labs' Mercury 3 diffusion LLM hits 1,000 tokens/sec

How we ranked this

Signal score
10 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Significant
New model release from a lab (Inception Labs) with a novel architecture (diffusion LLM) and performance claims. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Same-day
Cluster formed today. Ranking reflects the current source set at time of score.

Full methodology in our editorial standards.

COVERAGE [1]

  1. dev.to — LLM tag TIER_1 English(EN) · Prabhakar Chaudhary ·

    Mercury 3: How Inception Labs Scaled Diffusion LLMs to 1,000 Tokens Per Second

    <h1> Mercury 3: How Inception Labs Scaled Diffusion LLMs to 1,000 Tokens Per Second </h1> <p>When most people think of large language models, they picture the same basic loop: generate one token, append it to the context, generate the next. Autoregressive decoding has been the de…

RELATED ENTITIES

RELATED TOPICS