Model comparison
Gemma 3 27B vs GLM-5.2
GLM-5.2 is the stronger model overall, scoring 50.9 to 30.5 on the Noometry Index. Gemma 3 27B costs 21× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
GoogleSummary
- They share 27 benchmarks with published results for both. Gemma 3 27B scores higher in 0 categories and GLM-5.2 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5.2 leads 57.1 to 25.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 22.5% for Gemma 3 27B and 86.4% for GLM-5.2.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $1.40 / $4.40 for GLM-5.2.
- GLM-5.2 accepts more context: 1M tokens versus 131K.
Side by side
| Gemma 3 27B | GLM-5.2 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | |
| Noometry Index | 30.5 | 50.9 |
| Released | 2025-03-11 | 2026-06-13 |
| Weights | Open | Open |
| Context window | 131K | 1M |
| Max output | 8K | 131K |
| Input $ / M tokens | $0.08 | $1.40 |
| Output $ / M tokens | $0.16 | $4.40 |
| Results tracked | 43 | 50 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
Gemma 3 27B: 22.1 (#329), GLM-5.2: 50.9 (#37)
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| SciCode | 21.2% | 50.5% |
| LMArena Coding | 1322 | 1484 |
| SWE-bench Verified | — | 78.7% |
| DeepSWE | — | 43.8% |
| FrontierCode | — | 24.5% |
| Aider Polyglot | 4.9% | — |
| LMArena WebDev | — | 1604 |
| WeirdML | — | 70.1% |
| LiveBench Coding | 39.9% | — |
| ALE-Bench | — | 1,047 |
Agentic & Tool Use GLM-5.2 leads
Gemma 3 27B: 24.9 (#105), GLM-5.2: 31.6 (#63)
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 31.7% |
| GBAEval | — | 0% |
| Vending-Bench 2 | — | 8,314 |
Reasoning GLM-5.2 leads
Gemma 3 27B: 16.3 (#297), GLM-5.2: 42.1 (#46)
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| Kagi LLM Benchmark | 40.4% | 62.6% |
| CritPt | 0% | 20.9% |
| Chess Puzzles | 0% | 21% |
| LMArena Hard Prompts | 1340 | 1480 |
| DTBench | 52.5% | 93.6% |
| LMCA | 12.3% | 45.8% |
| Epoch Capabilities Index | 130.04 | 151.76 |
| ARC-AGI-2 | — | 22.8% |
| SimpleBench | — | 58.8% |
| NYT Connections (extended) | — | 74.3% |
| ARC-AGI-1 | — | 77% |
| EBR-Bench | — | 9.5% |
| LiveBench Reasoning | 43.8% | — |
| Mystery Game Puzzles | — | 19% |
| LiveBench Data Analysis | 51.5% | — |
| Surface Evolver Bench | — | 55.6% |
| LiveBench | 50% | — |
Math GLM-5.2 leads
Gemma 3 27B: 24.4 (#261), GLM-5.2: 55.4 (#35)
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 86.4% |
| LMArena Math | 1312 | 1482 |
| FrontierMath (Tiers 1-3) | — | 59.2% |
| FrontierMath Tier 4 | — | 29.3% |
| MathArena Final-Answer Competitions | — | 67.6% |
| ProofBench | — | 35% |
| LiveBench Math | 55.4% | — |
| MATH Level 5 | 74% | — |
Knowledge GLM-5.2 leads
Gemma 3 27B: 25.4 (#256), GLM-5.2: 57.1 (#36)
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| GPQA Diamond | 47.7% | 91.9% |
| LMArena Expert | 1303 | 1484 |
| SimpleQA Verified | — | 34.2% |
| Confabulations | 40.3% | — |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Not comparable
Gemma 3 27B: 32.0 (#94), GLM-5.2: —
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| LMArena Vision | 1165 | — |
| GeoBench | 52% | — |
Multilingual GLM-5.2 leads
Gemma 3 27B: 47.0 (#150), GLM-5.2: 56.0 (#25)
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1334 | 1459 |
| LMArena Chinese | 1347 | 1518 |
| LMArena French | 1368 | 1476 |
| LMArena German | 1362 | 1470 |
| LMArena Japanese | 1287 | 1451 |
| LMArena Korean | 1308 | 1445 |
| LMArena Russian | 1349 | 1467 |
| LMArena Spanish | 1348 | 1477 |
Instruction Following GLM-5.2 leads
Gemma 3 27B: 70.8 (#155), GLM-5.2: 77.1 (#32)
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1321 | 1465 |
| LiveBench Instruction Following | 74.9% | — |
Long Context GLM-5.2 leads
Gemma 3 27B: 27.6 (#288), GLM-5.2: 45.3 (#43)
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1333 | 1479 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference GLM-5.2 leads
Gemma 3 27B: 52.7 (#161), GLM-5.2: 70.7 (#20)
| Benchmark | Gemma 3 27B | GLM-5.2 |
|---|---|---|
| LMArena Text | 1358 | 1470 |
| LMArena Creative Writing | 1346 | 1465 |
| EQ-Bench Creative Writing | 1266 | 1757 |
| LMArena Multi-Turn | 1344 | 1468 |
| Short-Story Creative Writing | 79.9% | — |
| EQ-Bench 4 | — | 1222 |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than GLM-5.2?
GLM-5.2 is the stronger model overall, scoring 50.9 to 30.5 on the Noometry Index. Gemma 3 27B costs 21× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 27B or GLM-5.2?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.
Is Gemma 3 27B or GLM-5.2 better for coding?
GLM-5.2 scores higher on coding benchmarks: 50.9 versus 22.1 in the Noometry coding category.
Which has the bigger context window?
GLM-5.2 does, with 1M tokens against 131K.
How many benchmarks do Gemma 3 27B and GLM-5.2 share?
27 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and GLM-5.2 has 50.