Model comparison
GLM-5.3 vs Nemotron 3 Ultra
GLM-5.3 is the stronger model overall, scoring 54.8 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 2.3× less per token, which makes it the better buy when GLM-5.3's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
NVIDIASummary
- They share 30 benchmarks with published results for both. GLM-5.3 scores higher in 9 categories and Nemotron 3 Ultra in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.3 leads 62.3 to 35.3.
- The biggest single-benchmark swing is ProofBench: 49% for GLM-5.3 and 2% for Nemotron 3 Ultra.
- Nemotron 3 Ultra is cheaper at $0.50 / $2.20 per million input/output tokens, against $1.40 / $4.40 for GLM-5.3.
- GLM-5.3 accepts more context: 1M tokens versus 262K.
Side by side
| GLM-5.3 | Nemotron 3 Ultra | |
|---|---|---|
| Provider | Z.ai (Zhipu) | NVIDIA |
| Noometry Index | 54.8 | 42.5 |
| Released | 2026-08-14 | 2026-06-04 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 131K | 128K |
| Input $ / M tokens | $1.40 | $0.50 |
| Output $ / M tokens | $4.40 | $2.20 |
| Results tracked | 42 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3 leads
GLM-5.3: 59.5 (#14), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | GLM-5.3 | Nemotron 3 Ultra |
|---|---|---|
| FrontierCode | 40.1% | 13.6% |
| SciCode | 59% | 40.3% |
| WeirdML | 75.4% | 43.5% |
| LMArena Coding | 1496 | 1468 |
| DeepSWE | 69% | — |
| CursorBench | 42.6% | — |
| LMArena WebDev | 1622 | — |
| FrontierSWE | 30.2% | — |
| ALE-Bench | 1,317 | — |
Agentic & Tool Use GLM-5.3 leads
GLM-5.3: 36.4 (#38), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | GLM-5.3 | Nemotron 3 Ultra |
|---|---|---|
| APEX-Agents | 56.6% | 22.7% |
| Vending-Bench 2 | 8,164 | — |
Reasoning GLM-5.3 leads
GLM-5.3: 46.1 (#46), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | GLM-5.3 | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 19.1% | 3.1% |
| Chess Puzzles | 21% | 12% |
| LMArena Hard Prompts | 1489 | 1452 |
| Mystery Game Puzzles | 33% | 20% |
| DTBench | 87.7% | 90.1% |
| LMCA | 55.5% | 36.9% |
| Epoch Capabilities Index | 155.61 | 146.17 |
| NYT Connections (extended) | 74.2% | — |
| Bench to the Future 3 | 0.15 | — |
Math GLM-5.3 leads
GLM-5.3: 62.3 (#33), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | GLM-5.3 | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.1% | 86.7% |
| ProofBench | 49% | 2% |
| LMArena Math | 1489 | 1457 |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
Knowledge GLM-5.3 leads
GLM-5.3: 58.3 (#37), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | GLM-5.3 | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 90.9% | 85.4% |
| LMArena Expert | 1516 | 1472 |
| SimpleQA Verified | 41% | — |
Multilingual GLM-5.3 leads
GLM-5.3: 55.7 (#28), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | GLM-5.3 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1457 | 1427 |
| LMArena Chinese | 1528 | 1497 |
| LMArena French | 1499 | 1472 |
| LMArena German | 1499 | 1471 |
| LMArena Korean | 1472 | 1386 |
| LMArena Russian | 1463 | 1417 |
| LMArena Spanish | 1460 | 1454 |
| LMArena Japanese | 1453 | — |
Instruction Following GLM-5.3 leads
GLM-5.3: 77.5 (#23), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | GLM-5.3 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1477 | 1418 |
Long Context GLM-5.3 leads
GLM-5.3: 45.4 (#41), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | GLM-5.3 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1482 | 1437 |
Writing & Preference GLM-5.3 leads
GLM-5.3: 75.7 (#6), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | GLM-5.3 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1471 | 1445 |
| LMArena Creative Writing | 1457 | 1398 |
| EQ-Bench Creative Writing | 2075 | 1692 |
| LMArena Multi-Turn | 1472 | 1411 |
Frequently asked questions
Is GLM-5.3 better than Nemotron 3 Ultra?
GLM-5.3 is the stronger model overall, scoring 54.8 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 2.3× less per token, which makes it the better buy when GLM-5.3's lead doesn't matter for your workload.
Which is cheaper, GLM-5.3 or Nemotron 3 Ultra?
Nemotron 3 Ultra is cheaper. It lists at $0.50 per million input tokens and $2.20 per million output tokens; GLM-5.3 lists at $1.40 and $4.40.
Is GLM-5.3 or Nemotron 3 Ultra better for coding?
GLM-5.3 scores higher on coding benchmarks: 59.5 versus 38.1 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3 does, with 1M tokens against 262K.
How many benchmarks do GLM-5.3 and Nemotron 3 Ultra share?
30 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Nemotron 3 Ultra has 30.