Model comparison
Gemma 3 12B vs GLM-5
GLM-5 is the stronger model overall, scoring 46.1 to 32.1 on the Noometry Index. Gemma 3 12B costs 21× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
GoogleSummary
- They share 18 benchmarks with published results for both. Gemma 3 12B scores higher in 0 categories and GLM-5 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5 leads 52.3 to 26.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 16.7% for Gemma 3 12B and 80% for GLM-5.
- Gemma 3 12B is cheaper at $0.05 / $0.15 per million input/output tokens, against $1 / $3.20 for GLM-5.
- GLM-5 accepts more context: 205K tokens versus 131K.
Side by side
| Gemma 3 12B | GLM-5 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | |
| Noometry Index | 32.1 | 46.1 |
| Released | 2025-03-12 | 2026-02-11 |
| Weights | Open | Open |
| Context window | 131K | 205K |
| Max output | 8K | 131K |
| Input $ / M tokens | $0.05 | $1 |
| Output $ / M tokens | $0.15 | $3.20 |
| Results tracked | 24 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
Gemma 3 12B: 31.7 (#280), GLM-5: 49.0 (#52)
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| LMArena Coding | 1281 | 1461 |
| SWE-bench Verified | — | 72.1% |
| SWE-bench Verified (bash only) | — | 72.8% |
| LMArena WebDev | — | 1434 |
| SWE-bench Multilingual | — | 69.7% |
| SciCode | 17.4% | — |
| WeirdML | — | 48.2% |
| ALE-Bench | — | 765.62 |
Agentic & Tool Use GLM-5 leads
Gemma 3 12B: 25.5 (#108), GLM-5: 31.1 (#71)
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| Terminal-Bench | — | 52.4% |
| Berkeley Function Calling Leaderboard | 30.4% | — |
| τ²-bench Airline | — | 82.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 73.7% |
| τ²-bench Telecom | — | 86.8% |
| Vending-Bench 2 | — | 4,432 |
Reasoning GLM-5 leads
Gemma 3 12B: 15.7 (#313), GLM-5: 27.6 (#116)
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| Chess Puzzles | 0% | 10% |
| LMArena Hard Prompts | 1309 | 1452 |
| Epoch Capabilities Index | 123.5 | 145.83 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | — | 53.2% |
| Kagi LLM Benchmark | — | 75% |
| NYT Connections (extended) | — | 74.8% |
| ARC-AGI-1 | — | 44.7% |
| CritPt | 0% | — |
| DTBench | 48.8% | — |
| LMCA | 4.5% | — |
| ForecastBench | — | 61 |
Math GLM-5 leads
Gemma 3 12B: 22.3 (#279), GLM-5: 46.4 (#71)
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 16.7% | 80% |
| LMArena Math | 1307 | 1440 |
| MathArena Final-Answer Competitions | — | 65.7% |
| FrontierMath (Feb 2025 set) | — | 16.4% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GLM-5 leads
Gemma 3 12B: 26.5 (#257), GLM-5: 52.3 (#64)
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| GPQA Diamond | 39.5% | 87.8% |
| Vectara Hallucination Rate | 4.4% | 10.1% |
| LMArena Expert | 1248 | 1454 |
Multimodal Not comparable
Gemma 3 12B: —, GLM-5: —
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| MindCube | 46.7% | — |
Multilingual GLM-5 leads
Gemma 3 12B: 45.7 (#165), GLM-5: 53.7 (#58)
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| LMArena Non-English | 1318 | 1430 |
| LMArena German | 1370 | 1445 |
| LMArena Russian | 1335 | 1436 |
| LMArena Chinese | — | 1511 |
| LMArena French | — | 1455 |
| LMArena Japanese | — | 1416 |
| LMArena Korean | — | 1423 |
| LMArena Spanish | — | 1454 |
Instruction Following GLM-5 leads
Gemma 3 12B: 68.6 (#186), GLM-5: 75.2 (#67)
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| LMArena Instruction Following | 1299 | 1428 |
Long Context GLM-5 leads
Gemma 3 12B: 40.0 (#162), GLM-5: 44.7 (#60)
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| LMArena Longer Query | 1317 | 1446 |
| CL-bench | — | 18.7% |
Writing & Preference GLM-5 leads
Gemma 3 12B: 47.5 (#209), GLM-5: 66.0 (#38)
| Benchmark | Gemma 3 12B | GLM-5 |
|---|---|---|
| LMArena Text | 1334 | 1446 |
| LMArena Creative Writing | 1331 | 1439 |
| EQ-Bench Creative Writing | 1126 | 1601 |
| LMArena Multi-Turn | 1334 | 1456 |
Frequently asked questions
Is Gemma 3 12B better than GLM-5?
GLM-5 is the stronger model overall, scoring 46.1 to 32.1 on the Noometry Index. Gemma 3 12B costs 21× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 12B or GLM-5?
Gemma 3 12B is cheaper. It lists at $0.05 per million input tokens and $0.15 per million output tokens; GLM-5 lists at $1 and $3.20.
Is Gemma 3 12B or GLM-5 better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 31.7 in the Noometry coding category.
Which has the bigger context window?
GLM-5 does, with 205K tokens against 131K.
How many benchmarks do Gemma 3 12B and GLM-5 share?
18 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and GLM-5 has 45.