Compare / Nemotron Labs Dory 1.0 4B vs Qwen3 4B Thinking 2507
Nemotron Labs Dory 1.0 4B vs Qwen3 4B Thinking 2507
Updated November 2026
Verdict
Nemotron Labs Dory 1.0 4B scores higher on the Small LM Index: 60.0 vs 55.7, although their likely ranges overlap, so the lead is not certain. On the 1 benchmark both labs report, Nemotron Labs Dory 1.0 4B wins 1 and Qwen3 4B Thinking 2507 wins 0. Both need about 4.2 GB of memory at 4-bit quantization. Neither model has a hosted API on OpenRouter, so you would run both yourself.
Pick Nemotron Labs Dory 1.0 4B if…
- you want the strongest model overall
Pick Qwen3 4B Thinking 2507 if…
- you need a permissive license (Apache-2.0)
- you want the more widely used model (more quantized builds and community support)
Specifications
| Nemotron Labs Dory 1.0 4B | Qwen3 4B Thinking 2507 | |
|---|---|---|
| Small LM Index | 60.0 (likely 58–62) | 55.7 (likely 51–60) |
| Lab | NVIDIA · United States | Alibaba (Qwen) · China |
| Released | 2026-10-16 | 2025-08-06 |
| Parameters | 4B (dense) | 4B (dense) |
| Memory at 4-bit | ~3.4 GB | ~4.2 GB |
| Context window | 262K tokens | 262K tokens |
| Input | text | text |
| Reasoning | Always thinks | Always thinks |
| License | OpenMDW-1.1 | Apache-2.0 |
| API price / 1M tokens | Self-host only | Self-host only |
| Hugging Face downloads (30 days) | 10.5K | 440.9K |
Benchmarks
Higher is better. Bold marks the winner where both labs report a score. ≈ marks the index model's estimate for a score that wasn't reported, ✓ an independent measurement and * a score measured by another lab.
| Benchmark | Nemotron Labs Dory 1.0 4B | Qwen3 4B Thinking 2507 |
|---|---|---|
| MMLU-Pro Knowledge | ≈ 79 | 74.0 |
| GPQA Diamond Knowledge | 69.2 | 65.8 |
| Humanity's Last Exam Reasoning | 8.3 | ≈ 8 |
| AIME Math | ≈ 84 | 81.3 |
| LiveCodeBench Coding | ≈ 69 | 55.2 |
| SWE-bench Verified Coding | 50.3 | ≈ 39 |
| Terminal-Bench Agentic | 12.5 | — |
| IFBench Instruction following | 74.3 | — |
| AA-LCR Long context | 63.0 | — |
Scores are mostly reported by each lab with its own settings, so small gaps are noise. See the methodology. Open in the interactive comparison to add more models.
Questions
Which is better, Nemotron Labs Dory 1.0 4B or Qwen3 4B Thinking 2507?
Nemotron Labs Dory 1.0 4B scores higher on the Small LM Index: 60.0 vs 55.7, although their likely ranges overlap, so the lead is not certain. On the 1 benchmark both labs report, Nemotron Labs Dory 1.0 4B wins 1 and Qwen3 4B Thinking 2507 wins 0.
Which is cheaper to run?
Neither model has a hosted API on OpenRouter, so you would run both yourself. Both need about 4.2 GB of memory at 4-bit quantization.
Which has the longer context window?
Both have a 262K-token context window.
Can I use Nemotron Labs Dory 1.0 4B and Qwen3 4B Thinking 2507 commercially?
Nemotron Labs Dory 1.0 4B is released under OpenMDW-1.1 and Qwen3 4B Thinking 2507 under Apache-2.0. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.