Model comparison
GLM-5.3 vs Nemotron 3 Super
GLM-5.3 is the stronger model overall, scoring 54.6 to 39.6 on the Noometry Index. Nemotron 3 Super costs 12× less per token, which makes it the better buy when GLM-5.3's lead doesn't matter for your workload.
Last verified . 20 shared benchmarks.
NVIDIASummary
- They share 20 benchmarks with published results for both. GLM-5.3 scores higher in 8 categories and Nemotron 3 Super in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.3 leads 45.2 to 17.9.
- The biggest single-benchmark swing is NYT Connections (extended): 74.2% for GLM-5.3 and 15.4% for Nemotron 3 Super.
- Nemotron 3 Super is cheaper at $0.08 / $0.45 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 Super | |
|---|---|---|
| Provider | Z.ai (Zhipu) | NVIDIA |
| Noometry Index | 54.6 | 39.6 |
| Released | 2026-08-14 | 2026-03-11 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 131K | 131K |
| Input $ / M tokens | $1.40 | $0.08 |
| Output $ / M tokens | $4.40 | $0.45 |
| Results tracked | 41 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3 leads
GLM-5.3: 59.0 (#12), Nemotron 3 Super: 38.9 (#153)
| Benchmark | GLM-5.3 | Nemotron 3 Super |
|---|---|---|
| SciCode | 59% | 36% |
| WeirdML | 75.4% | 38% |
| LMArena Coding | 1499 | 1403 |
| ALE-Bench | 1,317 | 213.9 |
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| LMArena WebDev | 1623 | — |
| FrontierSWE | 30.2% | — |
Agentic & Tool Use Not comparable
GLM-5.3: 36.3 (#34), Nemotron 3 Super: —
| Benchmark | GLM-5.3 | Nemotron 3 Super |
|---|---|---|
| APEX-Agents | 56.6% | — |
| Vending-Bench 2 | 8,164 | — |
Reasoning GLM-5.3 leads
GLM-5.3: 45.2 (#42), Nemotron 3 Super: 17.9 (#274)
| Benchmark | GLM-5.3 | Nemotron 3 Super |
|---|---|---|
| NYT Connections (extended) | 74.2% | 15.4% |
| CritPt | 19.1% | 3.1% |
| LMArena Hard Prompts | 1491 | 1386 |
| Chess Puzzles | 21% | — |
| Mystery Game Puzzles | 33% | — |
| DTBench | 87.7% | — |
| LMCA | 55.5% | — |
| Bench to the Future 3 | 0.15 | — |
| Epoch Capabilities Index | 155.56 | — |
Math GLM-5.3 leads
GLM-5.3: 62.1 (#27), Nemotron 3 Super: 37.6 (#95)
| Benchmark | GLM-5.3 | Nemotron 3 Super |
|---|---|---|
| LMArena Math | 1493 | 1381 |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
| MathArena Final-Answer Competitions | — | 60.4% |
| OTIS Mock AIME 2024-2025 | 91.1% | — |
| ProofBench | 49% | — |
Knowledge GLM-5.3 leads
GLM-5.3: 58.3 (#34), Nemotron 3 Super: 38.2 (#136)
| Benchmark | GLM-5.3 | Nemotron 3 Super |
|---|---|---|
| LMArena Expert | 1512 | 1397 |
| GPQA Diamond | 90.9% | — |
| SimpleQA Verified | 41% | — |
Multilingual GLM-5.3 leads
GLM-5.3: 55.9 (#27), Nemotron 3 Super: 48.6 (#139)
| Benchmark | GLM-5.3 | Nemotron 3 Super |
|---|---|---|
| LMArena Non-English | 1457 | 1356 |
| LMArena Chinese | 1523 | 1430 |
| LMArena French | 1499 | 1404 |
| LMArena German | 1505 | 1341 |
| LMArena Russian | 1460 | 1336 |
| LMArena Spanish | 1444 | 1417 |
| LMArena Japanese | 1438 | — |
| LMArena Korean | 1478 | — |
Instruction Following GLM-5.3 leads
GLM-5.3: 77.8 (#15), Nemotron 3 Super: 71.4 (#149)
| Benchmark | GLM-5.3 | Nemotron 3 Super |
|---|---|---|
| LMArena Instruction Following | 1481 | 1347 |
Long Context GLM-5.3 leads
GLM-5.3: 45.5 (#32), Nemotron 3 Super: 41.5 (#134)
| Benchmark | GLM-5.3 | Nemotron 3 Super |
|---|---|---|
| LMArena Longer Query | 1485 | 1362 |
Writing & Preference GLM-5.3 leads
GLM-5.3: 75.9 (#6), Nemotron 3 Super: 56.1 (#135)
| Benchmark | GLM-5.3 | Nemotron 3 Super |
|---|---|---|
| LMArena Text | 1473 | 1378 |
| LMArena Creative Writing | 1459 | 1317 |
| LMArena Multi-Turn | 1473 | 1369 |
| EQ-Bench Creative Writing | 2075 | — |
Frequently asked questions
Is GLM-5.3 better than Nemotron 3 Super?
GLM-5.3 is the stronger model overall, scoring 54.6 to 39.6 on the Noometry Index. Nemotron 3 Super costs 12× 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 Super?
Nemotron 3 Super is cheaper. It lists at $0.08 per million input tokens and $0.45 per million output tokens; GLM-5.3 lists at $1.40 and $4.40.
Is GLM-5.3 or Nemotron 3 Super better for coding?
GLM-5.3 scores higher on coding benchmarks: 59.0 versus 38.9 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 Super share?
20 benchmarks have published results for both models. GLM-5.3 has 41 scored results on Noometry and Nemotron 3 Super has 21.