Skip to content
View original post on X: ModelScopeOfficial· Pick67/100AI score67/100

NVIDIA Nemotron Labs releases Dory1.0-4B, an open-weight reasoning model with 256K context

AISummary

NVIDIA Nemotron Labs released Dory1.0-4B, a 4B-parameter open-weight reasoning model with up to 256K context. Developers can run its shared recurrent block for different numbers of loops to balance latency and accuracy without retraining. The vendor reports 59.5 on SWE-bench Verified at two loops, versus 54.7 for Qwen3.5-4B.

AIWhy it matters

The post explains how one checkpoint trades latency for accuracy through recurrent loop count, a configuration choice that matters for deployment planning.

Post on XView on X
ModelScopeVerified on X
@ModelScope2022

NVIDIA Nemotron Labs Dory1.0-4B is an open-weight reasoning model with up to 256K context, letting developers adjust inference compute within a single checkpoint.
🤖 https://modelscope.ai/models/nv-community/NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16

🔁 One checkpoint, adjustable compute: Run the shared recurrent block for different numbers of loops to balance latency and accuracy per task, with no retraining or model swap.
🏆 Stronger coding-agent results at 4B: At two loops, it reaches 59.5 on SWE-bench Verified in the vendor evaluation, ahead of Qwen3.5-4B at 54.7.
🧠 Extra thinking without longer answers: More loops add latent compute per token instead of simply generating more tokens.
⚡ Loop without compounding cache cost: Optional last-loop KV caching keeps one cache per recurrent attention layer, avoiding extra KV-cache overhead as loop count grows.
📚 Built for long-context workloads: 4B parameters, up to 256K context, and releases in BF16, MXFP8, NVFP4, and Q5_K_M formats.
📜 OpenMDW License Agreement v1.1.

Source: ModelScope · x.comPublished · added here

Topics

NVIDIAAI agentsAI codingReasoningOpen-source ecosystemModel releasesEvals & benchmarks