Model comparison
GLM-4.6 vs Llama 3-70B
GLM-4.6 is the stronger model overall, scoring 41.4 to 28.8 on the Noometry Index.
Last verified . 18 shared benchmarks.
MetaSummary
- They share 18 benchmarks with published results for both. GLM-4.6 scores higher in 9 categories and Llama 3-70B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-4.6 leads 39.1 to 12.8.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 47.4% for GLM-4.6 and 35.1% for Llama 3-70B.
Side by side
| GLM-4.6 | Llama 3-70B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 41.4 | 28.8 |
| Released | 2025-09-30 | 2024-04-18 |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.60 | — |
| Output $ / M tokens | $2.20 | — |
| Results tracked | 29 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-4.6 leads
GLM-4.6: 40.1 (#148), Llama 3-70B: 35.8 (#218)
| Benchmark | GLM-4.6 | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1449 | 1206 |
| SWE-bench Verified (bash only) | 55.4% | — |
| LMArena WebDev | 1340 | — |
| SciCode | 38.4% | — |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| ALE-Bench | 340.82 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use GLM-4.6 leads
GLM-4.6: 32.3 (#66), Llama 3-70B: 21.1 (#139)
| Benchmark | GLM-4.6 | Llama 3-70B |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |
| Cybench | — | 5% |
Reasoning GLM-4.6 leads
GLM-4.6: 23.7 (#172), Llama 3-70B: 18.0 (#288)
| Benchmark | GLM-4.6 | Llama 3-70B |
|---|---|---|
| Kagi LLM Benchmark | 47.4% | 35.1% |
| LMArena Hard Prompts | 1440 | 1195 |
| CritPt | 1.1% | — |
| DTBench | — | 54.2% |
| Epoch Capabilities Index | — | 122.93 |
| ForecastBench | — | 57.1 |
| WinoGrande | — | 83.5% |
Math GLM-4.6 leads
GLM-4.6: 39.1 (#111), Llama 3-70B: 12.8 (#305)
| Benchmark | GLM-4.6 | Llama 3-70B |
|---|---|---|
| LMArena Math | 1432 | 1218 |
| OTIS Mock AIME 2024-2025 | — | 4.3% |
| MATH Level 5 | — | 22.6% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-4.6 leads
GLM-4.6: 40.2 (#124), Llama 3-70B: 20.8 (#277)
| Benchmark | GLM-4.6 | Llama 3-70B |
|---|---|---|
| LMArena Expert | 1431 | 1149 |
| GPQA Diamond | — | 40.6% |
| Vectara Hallucination Rate | 9.5% | — |
| MMLU | — | 79.3% |
Multilingual GLM-4.6 leads
GLM-4.6: 53.5 (#66), Llama 3-70B: 33.6 (#251)
| Benchmark | GLM-4.6 | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1426 | 1142 |
| LMArena Chinese | 1499 | 1114 |
| LMArena French | 1459 | 1232 |
| LMArena German | 1447 | 1169 |
| LMArena Japanese | 1393 | 1017 |
| LMArena Korean | 1400 | 1017 |
| LMArena Russian | 1419 | 1159 |
| LMArena Spanish | 1436 | 1241 |
Instruction Following GLM-4.6 leads
GLM-4.6: 74.3 (#98), Llama 3-70B: 62.5 (#238)
| Benchmark | GLM-4.6 | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1410 | 1194 |
Long Context GLM-4.6 leads
GLM-4.6: 43.4 (#94), Llama 3-70B: 35.6 (#240)
| Benchmark | GLM-4.6 | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1422 | 1174 |
Writing & Preference GLM-4.6 leads
GLM-4.6: 61.1 (#90), Llama 3-70B: 42.8 (#231)
| Benchmark | GLM-4.6 | Llama 3-70B |
|---|---|---|
| LMArena Text | 1440 | 1221 |
| LMArena Creative Writing | 1411 | 1210 |
| LMArena Multi-Turn | 1427 | 1223 |
| EQ-Bench Creative Writing | 1411 | — |
Frequently asked questions
Is GLM-4.6 better than Llama 3-70B?
GLM-4.6 is the stronger model overall, scoring 41.4 to 28.8 on the Noometry Index.
Is GLM-4.6 or Llama 3-70B better for coding?
GLM-4.6 scores higher on coding benchmarks: 40.1 versus 35.8 in the Noometry coding category.
How many benchmarks do GLM-4.6 and Llama 3-70B share?
18 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Llama 3-70B has 31.