Muse Spark 1.3 (xhigh) vs NVIDIA Nemotron Labs Dory 1.0
A direct, empirical evaluation of Muse Spark 1.3 (xhigh) (Meta AI) against NVIDIA Nemotron Labs Dory 1.0 (NVIDIA).
Muse Spark 1.3 (xhigh)
Developed by Meta AI · Proprietary Multimodal parameters
Autonomous codebase refactoring, terminal execution, million-token long context retrieval, and agentic workflows.
NVIDIA Nemotron Labs Dory 1.0
Developed by NVIDIA · 4 Billion Parameters (Looped Transformer) parameters
Ultra-low VRAM memory residency (4.4 GB FP8), 256k long-context needle retrieval, and recurrent layer depth execution.
| Metric | Muse Spark 1.3 (xhigh) | NVIDIA Nemotron Labs Dory 1.0 |
|---|---|---|
| SWE-bench Verified | 75.4% | 66.4% |
| MMLU-Pro | 88.4% | 82.8% |
| MATH-500 | 94.8% | 85.6% |
| Context Window | 1,000,000 tokens | 256,000 tokens |
| Input Pricing (per 1M) | $1.25 | $0.15 |
| Output Pricing (per 1M) | $5.00 | $0.45 |
Architectural & Practical Deployment Analysis
Evaluating Muse Spark 1.3 (xhigh) (Meta AI) alongside NVIDIA Nemotron Labs Dory 1.0 (NVIDIA) highlights key technical trade-offs across neural parameter scale and compute efficiency. Muse Spark 1.3 (xhigh) executes across Proprietary Multimodal total parameters with Dense Reasoner active per token, whereas NVIDIA Nemotron Labs Dory 1.0 operates on 4 Billion Parameters (Looped Transformer) parameters with 4.1B Active Recurrent Block (Equivalent to 16B Depth) active per token.
On SWE-bench Verified software engineering tasks, Muse Spark 1.3 (xhigh) achieves 75.4% autonomous resolution rate, compared to 66.4% for NVIDIA Nemotron Labs Dory 1.0. On complex reasoning benchmarks, Muse Spark 1.3 (xhigh) posts 88.4% on MMLU-Pro and 94.8% on MATH-500, against 82.8% and 85.6% respectively for NVIDIA Nemotron Labs Dory 1.0.
From an operational economics perspective, processing 10,000,000 mixed tokens (7M input / 3M output) costs approximately $23.75 on Muse Spark 1.3 (xhigh) versus $2.40 on NVIDIA Nemotron Labs Dory 1.0.
Recommendation: Deploy Muse Spark 1.3 (xhigh) when your infrastructure prioritizes autonomous codebase refactoring, terminal execution, million-token long context retrieval, and agentic workflows.. Deploy NVIDIA Nemotron Labs Dory 1.0 when workload requirements favor ultra-low vram memory residency (4.4 gb fp8), 256k long-context needle retrieval, and recurrent layer depth execution..