Model comparison
GLM-5.3 vs Llama 3.2 90B
GLM-5.3 is the stronger model overall, scoring 54.8 to 27.5 on the Noometry Index.
Last verified . 3 shared benchmarks.
MetaSummary
- They share 3 benchmarks with published results for both. GLM-5.3 scores higher in 4 categories and Llama 3.2 90B in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.3 leads 62.3 to 11.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 91.1% for GLM-5.3 and 2.6% for Llama 3.2 90B.
Side by side
| GLM-5.3 | Llama 3.2 90B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 54.8 | 27.5 |
| Released | 2026-08-14 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 131K | — |
| Input $ / M tokens | $1.40 | — |
| Output $ / M tokens | $4.40 | — |
| Results tracked | 42 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
GLM-5.3: 59.5 (#14), Llama 3.2 90B: —
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| CursorBench | 42.6% | — |
| LMArena WebDev | 1622 | — |
| FrontierSWE | 30.2% | — |
| SciCode | 59% | — |
| WeirdML | 75.4% | — |
| LMArena Coding | 1496 | — |
| ALE-Bench | 1,317 | — |
Agentic & Tool Use GLM-5.3 leads
GLM-5.3: 36.4 (#38), Llama 3.2 90B: 30.0 (#80)
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| APEX-Agents | 56.6% | — |
| BALROG | — | 27.3% |
| Vending-Bench 2 | 8,164 | — |
Reasoning GLM-5.3 leads
GLM-5.3: 46.1 (#46), Llama 3.2 90B: 21.7 (#217)
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| Epoch Capabilities Index | 155.61 | 125.5 |
| NYT Connections (extended) | 74.2% | — |
| CritPt | 19.1% | — |
| Chess Puzzles | 21% | — |
| EnigmaEval | — | 0.4% |
| LMArena Hard Prompts | 1489 | — |
| Mystery Game Puzzles | 33% | — |
| DTBench | 87.7% | — |
| LMCA | 55.5% | — |
| Bench to the Future 3 | 0.15 | — |
Math GLM-5.3 leads
GLM-5.3: 62.3 (#33), Llama 3.2 90B: 11.1 (#308)
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.1% | 2.6% |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
| ProofBench | 49% | — |
| LMArena Math | 1489 | — |
| MATH Level 5 | — | 39.4% |
Knowledge GLM-5.3 leads
GLM-5.3: 58.3 (#37), Llama 3.2 90B: 21.7 (#274)
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| GPQA Diamond | 90.9% | 41% |
| SimpleQA Verified | 41% | — |
| LMArena Expert | 1516 | — |
| MMLU | — | 80.3% |
Multimodal Not comparable
GLM-5.3: —, Llama 3.2 90B: 25.4 (#124)
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| LMArena Vision | — | 1000 |
| GeoBench | — | 52% |
Multilingual Not comparable
GLM-5.3: 55.7 (#28), Llama 3.2 90B: —
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| LMArena Non-English | 1457 | — |
| LMArena Chinese | 1528 | — |
| LMArena French | 1499 | — |
| LMArena German | 1499 | — |
| LMArena Japanese | 1453 | — |
| LMArena Korean | 1472 | — |
| LMArena Russian | 1463 | — |
| LMArena Spanish | 1460 | — |
Instruction Following Not comparable
GLM-5.3: 77.5 (#23), Llama 3.2 90B: —
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| LMArena Instruction Following | 1477 | — |
Long Context Not comparable
GLM-5.3: 45.4 (#41), Llama 3.2 90B: —
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| LMArena Longer Query | 1482 | — |
Writing & Preference Not comparable
GLM-5.3: 75.7 (#6), Llama 3.2 90B: —
| Benchmark | GLM-5.3 | Llama 3.2 90B |
|---|---|---|
| LMArena Text | 1471 | — |
| LMArena Creative Writing | 1457 | — |
| EQ-Bench Creative Writing | 2075 | — |
| LMArena Multi-Turn | 1472 | — |
Frequently asked questions
Is GLM-5.3 better than Llama 3.2 90B?
GLM-5.3 is the stronger model overall, scoring 54.8 to 27.5 on the Noometry Index.
How many benchmarks do GLM-5.3 and Llama 3.2 90B share?
3 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Llama 3.2 90B has 9.