Model comparison
Gemma 2 27B vs GLM-5.3-Flash
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 29.4 on the Noometry Index.
Last verified . 20 shared benchmarks.
Summary
- They share 20 benchmarks with published results for both. Gemma 2 27B scores higher in 0 categories and GLM-5.3-Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.3-Flash leads 53.3 to 10.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.4% for Gemma 2 27B and 93.9% for GLM-5.3-Flash.
- GLM-5.3-Flash is cheaper at $0.15 / $0.50 per million input/output tokens, against $0.65 / $0.65 for Gemma 2 27B.
- GLM-5.3-Flash accepts more context: 1M tokens versus 8K.
Side by side
| Gemma 2 27B | GLM-5.3-Flash | |
|---|---|---|
| Provider | Z.ai (Zhipu) | |
| Noometry Index | 29.4 | 51.8 |
| Released | 2024-06-24 | 2026-08-20 |
| Weights | Open | Open |
| Context window | 8K | 1M |
| Max output | 2K | 131K |
| Input $ / M tokens | $0.65 | $0.15 |
| Output $ / M tokens | $0.65 | $0.50 |
| Results tracked | 34 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3-Flash leads
Gemma 2 27B: 34.1 (#246), GLM-5.3-Flash: 53.1 (#31)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| LMArena Coding | 1211 | 1508 |
| DeepSWE | — | 63.4% |
| FrontierCode | — | 31.8% |
| CursorBench | — | 36.8% |
| LMArena WebDev | — | 1609 |
| FrontierSWE | — | 18.1% |
| SciCode | — | 51.6% |
| BigCodeBench Instruct | 42.8% | — |
| LiveBench Coding | 36% | — |
| BigCodeBench Complete | 52.5% | — |
| ALE-Bench | — | 303.55 |
Agentic & Tool Use Not comparable
Gemma 2 27B: —, GLM-5.3-Flash: 34.2 (#47)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| APEX-Agents | — | 52.8% |
| GDP.pdf | — | 14% |
Reasoning GLM-5.3-Flash leads
Gemma 2 27B: 15.3 (#315), GLM-5.3-Flash: 48.0 (#42)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| LMArena Hard Prompts | 1198 | 1491 |
| Epoch Capabilities Index | 122.08 | 151.88 |
| ARC-AGI-2 | — | 65.8% |
| ARC-AGI-1 | — | 91% |
| CritPt | — | 15.4% |
| Chess Puzzles | — | 14% |
| LiveBench Reasoning | 28.1% | — |
| Mystery Game Puzzles | — | 8% |
| DTBench | 48% | — |
| LiveBench Data Analysis | 47.9% | — |
| LMCA | 7.1% | — |
| Surface Evolver Bench | — | 52.5% |
| Bench to the Future 3 | — | 0.15 |
| LiveBench | 38.2% | — |
Math GLM-5.3-Flash leads
Gemma 2 27B: 10.7 (#311), GLM-5.3-Flash: 53.3 (#47)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 93.9% |
| LMArena Math | 1212 | 1500 |
| FrontierMath (Tiers 1-3) | — | 55.8% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 21% |
| LiveBench Math | 26.5% | — |
| MATH Level 5 | 27.9% | — |
Knowledge GLM-5.3-Flash leads
Gemma 2 27B: 19.0 (#280), GLM-5.3-Flash: 58.4 (#36)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| GPQA Diamond | 36.5% | 90.2% |
| LMArena Expert | 1172 | 1513 |
| Confabulations | 27.1% | — |
| MMLU | 75.7% | — |
Multimodal Not comparable
Gemma 2 27B: —, GLM-5.3-Flash: 42.8 (#27)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| LMArena Vision | — | 1296 |
Multilingual GLM-5.3-Flash leads
Gemma 2 27B: 38.6 (#226), GLM-5.3-Flash: 56.0 (#25)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| LMArena Non-English | 1217 | 1462 |
| LMArena Chinese | 1221 | 1527 |
| LMArena French | 1247 | 1496 |
| LMArena German | 1209 | 1470 |
| LMArena Japanese | 1175 | 1429 |
| LMArena Korean | 1174 | 1446 |
| LMArena Russian | 1234 | 1469 |
| LMArena Spanish | 1228 | 1471 |
Instruction Following GLM-5.3-Flash leads
Gemma 2 27B: 60.5 (#249), GLM-5.3-Flash: 77.5 (#20)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| LMArena Instruction Following | 1206 | 1478 |
| LiveBench Instruction Following | 58.1% | — |
Long Context GLM-5.3-Flash leads
Gemma 2 27B: 37.3 (#218), GLM-5.3-Flash: 45.4 (#39)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| LMArena Longer Query | 1231 | 1482 |
Writing & Preference GLM-5.3-Flash leads
Gemma 2 27B: 44.2 (#225), GLM-5.3-Flash: 65.3 (#50)
| Benchmark | Gemma 2 27B | GLM-5.3-Flash |
|---|---|---|
| LMArena Text | 1231 | 1471 |
| LMArena Creative Writing | 1241 | 1442 |
| LMArena Multi-Turn | 1224 | 1467 |
| LiveBench Language | 32.6% | — |
Frequently asked questions
Is Gemma 2 27B better than GLM-5.3-Flash?
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 29.4 on the Noometry Index.
Which is cheaper, Gemma 2 27B or GLM-5.3-Flash?
GLM-5.3-Flash is cheaper. It lists at $0.15 per million input tokens and $0.50 per million output tokens; Gemma 2 27B lists at $0.65 and $0.65.
Is Gemma 2 27B or GLM-5.3-Flash better for coding?
GLM-5.3-Flash scores higher on coding benchmarks: 53.1 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3-Flash does, with 1M tokens against 8K.
How many benchmarks do Gemma 2 27B and GLM-5.3-Flash share?
20 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and GLM-5.3-Flash has 40.