Model comparison
GLM-5.2 vs Llama 4 Maverick
GLM-5.2 is the stronger model overall, scoring 51.1 to 30.9 on the Noometry Index. Llama 4 Maverick costs 7.1× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
MetaSummary
- They share 32 benchmarks with published results for both. GLM-5.2 scores higher in 9 categories and Llama 4 Maverick in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.3 to 10.1.
- The biggest single-benchmark swing is ARC-AGI-1: 77% for GLM-5.2 and 4.4% for Llama 4 Maverick.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $1.40 / $4.40 for GLM-5.2.
- GLM-5.2 accepts more context: 1M tokens versus 128K.
Side by side
| GLM-5.2 | Llama 4 Maverick | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 51.1 | 30.9 |
| Released | 2026-06-13 | 2025-04-05 |
| Weights | Open | Open |
| Context window | 1M | 128K |
| Max output | 131K | 4K |
| Input $ / M tokens | $1.40 | $0.19 |
| Output $ / M tokens | $4.40 | $0.65 |
| Results tracked | 51 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
GLM-5.2: 51.3 (#41), Llama 4 Maverick: 26.6 (#324)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| SciCode | 50.5% | 33.1% |
| WeirdML | 70.1% | 24.5% |
| LMArena Coding | 1485 | 1302 |
| ALE-Bench | 1,047 | 172.97 |
| SWE-bench Verified | 78.7% | — |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| SWE-bench Verified (bash only) | — | 21% |
| Aider Polyglot | — | 15.6% |
| LMArena WebDev | 1603 | — |
| BigCodeBench Instruct | — | 49.7% |
| BigCodeBench Complete | — | 61.4% |
Agentic & Tool Use GLM-5.2 leads
GLM-5.2: 32.4 (#63), Llama 4 Maverick: 28.2 (#91)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| APEX-Agents | 45.2% | — |
| Berkeley Function Calling Leaderboard | — | 37.3% |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 8,314 | — |
Reasoning GLM-5.2 leads
GLM-5.2: 42.3 (#52), Llama 4 Maverick: 10.1 (#342)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| ARC-AGI-2 | 22.8% | 0% |
| SimpleBench | 58.8% | 27.7% |
| Kagi LLM Benchmark | 62.6% | 55.9% |
| NYT Connections (extended) | 74.3% | 8% |
| ARC-AGI-1 | 77% | 4.4% |
| CritPt | 20.9% | 0% |
| LMArena Hard Prompts | 1480 | 1281 |
| DTBench | 93.6% | 61.9% |
| LMCA | 45.8% | 15.9% |
| Epoch Capabilities Index | 151.78 | 132.2 |
| Chess Puzzles | 21% | — |
| EnigmaEval | — | 0.6% |
| EBR-Bench | 9.5% | — |
| Mystery Game Puzzles | 19% | — |
| Surface Evolver Bench | 55.6% | — |
| ForecastBench | — | 57.5 |
Math GLM-5.2 leads
GLM-5.2: 55.7 (#43), Llama 4 Maverick: 26.0 (#262)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.4% | 20.6% |
| LMArena Math | 1482 | 1299 |
| FrontierMath (Tiers 1-3) | 59.2% | — |
| FrontierMath Tier 4 | 29.3% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| ProofBench | 35% | — |
| Omni-MATH | — | 42.2% |
| MATH Level 5 | — | 73% |
| FrontierMath (Feb 2025 set) | — | 0.7% |
Knowledge GLM-5.2 leads
GLM-5.2: 57.1 (#40), Llama 4 Maverick: 33.4 (#204)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 91.9% | 67% |
| LMArena Expert | 1486 | 1259 |
| Humanity's Last Exam | — | 5.7% |
| SimpleQA Verified | 34.2% | — |
| MMLU-Pro | — | 81% |
| Confabulations | — | 22.6% |
| Vectara Hallucination Rate | — | 8.2% |
| GPQA (HELM) | — | 65% |
Multimodal Not comparable
GLM-5.2: —, Llama 4 Maverick: 31.6 (#105)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | — | 1142 |
| GeoBench | — | 52% |
| SpatialViz-Bench | — | 31.8% |
Multilingual GLM-5.2 leads
GLM-5.2: 55.8 (#26), Llama 4 Maverick: 42.2 (#195)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1459 | 1269 |
| LMArena Chinese | 1519 | 1277 |
| LMArena French | 1479 | 1259 |
| LMArena German | 1468 | 1291 |
| LMArena Japanese | 1451 | 1207 |
| LMArena Korean | 1445 | 1203 |
| LMArena Russian | 1466 | 1286 |
| LMArena Spanish | 1477 | 1293 |
Instruction Following GLM-5.2 leads
GLM-5.2: 76.9 (#34), Llama 4 Maverick: 71.7 (#146)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1465 | 1267 |
| IFEval | — | 90.8% |
Long Context GLM-5.2 leads
GLM-5.2: 45.3 (#43), Llama 4 Maverick: 31.4 (#279)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1479 | 1280 |
| Fiction.LiveBench | — | 46.2% |
Writing & Preference GLM-5.2 leads
GLM-5.2: 70.4 (#21), Llama 4 Maverick: 38.8 (#252)
| Benchmark | GLM-5.2 | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1470 | 1287 |
| LMArena Creative Writing | 1462 | 1267 |
| EQ-Bench Creative Writing | 1757 | 860 |
| LMArena Multi-Turn | 1469 | 1289 |
| Short-Story Creative Writing | — | 62% |
| WildBench | — | 80% |
| EQ-Bench 4 | 1222 | — |
Frequently asked questions
Is GLM-5.2 better than Llama 4 Maverick?
GLM-5.2 is the stronger model overall, scoring 51.1 to 30.9 on the Noometry Index. Llama 4 Maverick costs 7.1× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Which is cheaper, GLM-5.2 or Llama 4 Maverick?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.
Is GLM-5.2 or Llama 4 Maverick better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
GLM-5.2 does, with 1M tokens against 128K.
How many benchmarks do GLM-5.2 and Llama 4 Maverick share?
32 benchmarks have published results for both models. GLM-5.2 has 51 scored results on Noometry and Llama 4 Maverick has 54.