Model comparison
GLM-5.2 vs Llama 3.1-70B
GLM-5.2 is the stronger model overall, scoring 51.1 to 29.6 on the Noometry Index. Llama 3.1-70B costs 5.4× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
MetaSummary
- They share 24 benchmarks with published results for both. GLM-5.2 scores higher in 9 categories and Llama 3.1-70B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.2 leads 55.7 to 13.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 86.4% for GLM-5.2 and 3.6% for Llama 3.1-70B.
- Llama 3.1-70B is cheaper at $0.40 / $0.40 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 3.1-70B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 51.1 | 29.6 |
| Released | 2026-06-13 | 2024-07-23 |
| Weights | Open | Open |
| Context window | 1M | 128K |
| Max output | 131K | 4K |
| Input $ / M tokens | $1.40 | $0.40 |
| Output $ / M tokens | $4.40 | $0.40 |
| Results tracked | 51 | 35 |
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 3.1-70B: 30.3 (#296)
| Benchmark | GLM-5.2 | Llama 3.1-70B |
|---|---|---|
| WeirdML | 70.1% | 9% |
| LMArena Coding | 1485 | 1260 |
| SWE-bench Verified | 78.7% | — |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| LMArena WebDev | 1603 | — |
| SciCode | 50.5% | — |
| BigCodeBench Instruct | — | 46.1% |
| BigCodeBench Complete | — | 54.8% |
| ALE-Bench | 1,047 | — |
Agentic & Tool Use GLM-5.2 leads
GLM-5.2: 32.4 (#63), Llama 3.1-70B: 25.1 (#112)
| Benchmark | GLM-5.2 | Llama 3.1-70B |
|---|---|---|
| APEX-Agents | 45.2% | — |
| TheAgentCompany | — | 6.9% |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 31.7% | — |
| BALROG | — | 27.9% |
| GBAEval | 0% | — |
| Vending-Bench 2 | 8,314 | — |
Reasoning GLM-5.2 leads
GLM-5.2: 42.3 (#52), Llama 3.1-70B: 21.6 (#220)
| Benchmark | GLM-5.2 | Llama 3.1-70B |
|---|---|---|
| LMArena Hard Prompts | 1480 | 1241 |
| DTBench | 93.6% | 60% |
| LMCA | 45.8% | 14.8% |
| Epoch Capabilities Index | 151.78 | 125.92 |
| ARC-AGI-2 | 22.8% | — |
| SimpleBench | 58.8% | — |
| Kagi LLM Benchmark | 62.6% | — |
| NYT Connections (extended) | 74.3% | — |
| ARC-AGI-1 | 77% | — |
| CritPt | 20.9% | — |
| Chess Puzzles | 21% | — |
| EBR-Bench | 9.5% | — |
| Mystery Game Puzzles | 19% | — |
| Surface Evolver Bench | 55.6% | — |
Math GLM-5.2 leads
GLM-5.2: 55.7 (#43), Llama 3.1-70B: 13.5 (#304)
| Benchmark | GLM-5.2 | Llama 3.1-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.4% | 3.6% |
| LMArena Math | 1482 | 1252 |
| FrontierMath (Tiers 1-3) | 59.2% | — |
| FrontierMath Tier 4 | 29.3% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| ProofBench | 35% | — |
| Omni-MATH | — | 21% |
| MATH Level 5 | — | 36.7% |
Knowledge GLM-5.2 leads
GLM-5.2: 57.1 (#40), Llama 3.1-70B: 24.2 (#269)
| Benchmark | GLM-5.2 | Llama 3.1-70B |
|---|---|---|
| GPQA Diamond | 91.9% | 44.2% |
| LMArena Expert | 1486 | 1209 |
| SimpleQA Verified | 34.2% | — |
| MMLU-Pro | — | 65.3% |
| GPQA (HELM) | — | 42.6% |
| MMLU | — | 80.1% |
Multilingual GLM-5.2 leads
GLM-5.2: 55.8 (#26), Llama 3.1-70B: 38.8 (#225)
| Benchmark | GLM-5.2 | Llama 3.1-70B |
|---|---|---|
| LMArena Non-English | 1459 | 1219 |
| LMArena Chinese | 1519 | 1215 |
| LMArena French | 1479 | 1261 |
| LMArena German | 1468 | 1222 |
| LMArena Japanese | 1451 | 1132 |
| LMArena Korean | 1445 | 1140 |
| LMArena Russian | 1466 | 1234 |
| LMArena Spanish | 1477 | 1253 |
Instruction Following GLM-5.2 leads
GLM-5.2: 76.9 (#34), Llama 3.1-70B: 65.3 (#223)
| Benchmark | GLM-5.2 | Llama 3.1-70B |
|---|---|---|
| LMArena Instruction Following | 1465 | 1231 |
| IFEval | — | 82.1% |
Long Context GLM-5.2 leads
GLM-5.2: 45.3 (#43), Llama 3.1-70B: 37.6 (#214)
| Benchmark | GLM-5.2 | Llama 3.1-70B |
|---|---|---|
| LMArena Longer Query | 1479 | 1241 |
Writing & Preference GLM-5.2 leads
GLM-5.2: 70.4 (#21), Llama 3.1-70B: 35.4 (#267)
| Benchmark | GLM-5.2 | Llama 3.1-70B |
|---|---|---|
| LMArena Text | 1470 | 1261 |
| LMArena Creative Writing | 1462 | 1232 |
| EQ-Bench Creative Writing | 1757 | 784 |
| LMArena Multi-Turn | 1469 | 1256 |
| WildBench | — | 75.8% |
| EQ-Bench 4 | 1222 | — |
Frequently asked questions
Is GLM-5.2 better than Llama 3.1-70B?
GLM-5.2 is the stronger model overall, scoring 51.1 to 29.6 on the Noometry Index. Llama 3.1-70B costs 5.4× 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 3.1-70B?
Llama 3.1-70B is cheaper. It lists at $0.40 per million input tokens and $0.40 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.
Is GLM-5.2 or Llama 3.1-70B better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 30.3 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 3.1-70B share?
24 benchmarks have published results for both models. GLM-5.2 has 51 scored results on Noometry and Llama 3.1-70B has 35.