Model comparison
GLM-4.6 vs Llama 2-70B
GLM-4.6 is the stronger model overall, scoring 40.9 to 24.1 on the Noometry Index.
Last verified . 17 shared benchmarks.
MetaSummary
- They share 17 benchmarks with published results for both. GLM-4.6 scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-4.6 leads 39.6 to 7.4.
Side by side
| GLM-4.6 | Llama 2-70B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 40.9 | 24.1 |
| Released | 2025-09-30 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.60 | — |
| Output $ / M tokens | $2.20 | — |
| Results tracked | 29 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-4.6 leads
GLM-4.6: 39.9 (#137), Llama 2-70B: 30.8 (#281)
| Benchmark | GLM-4.6 | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1449 | 1079 |
| SWE-bench Verified (bash only) | 55.4% | — |
| LMArena WebDev | 1339 | — |
| SciCode | 38.4% | — |
| ALE-Bench | 340.82 | — |
Agentic & Tool Use Not comparable
GLM-4.6: 32.1 (#60), Llama 2-70B: —
| Benchmark | GLM-4.6 | Llama 2-70B |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |
Reasoning GLM-4.6 leads
GLM-4.6: 23.1 (#168), Llama 2-70B: 13.7 (#324)
| Benchmark | GLM-4.6 | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1440 | 1073 |
| Kagi LLM Benchmark | 47.4% | — |
| CritPt | 1.1% | — |
| DTBench | — | 41.6% |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| Epoch Capabilities Index | — | 113.59 |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math GLM-4.6 leads
GLM-4.6: 36.4 (#115), Llama 2-70B: 7.5 (#320)
| Benchmark | GLM-4.6 | Llama 2-70B |
|---|---|---|
| LMArena Math | 1431 | 1091 |
| OTIS Mock AIME 2024-2025 | — | 0% |
| MATH Level 5 | — | 3.3% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
| GSM8K | — | 69.6% |
Knowledge GLM-4.6 leads
GLM-4.6: 39.6 (#118), Llama 2-70B: 7.4 (#305)
| Benchmark | GLM-4.6 | Llama 2-70B |
|---|---|---|
| LMArena Expert | 1432 | 1039 |
| GPQA Diamond | — | 26.3% |
| Vectara Hallucination Rate | 9.5% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multilingual GLM-4.6 leads
GLM-4.6: 53.6 (#62), Llama 2-70B: 27.9 (#269)
| Benchmark | GLM-4.6 | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1426 | 1045 |
| LMArena Chinese | 1500 | 994 |
| LMArena French | 1459 | 1089 |
| LMArena German | 1447 | 1041 |
| LMArena Japanese | 1394 | 927 |
| LMArena Korean | 1401 | 964 |
| LMArena Russian | 1418 | 1083 |
| LMArena Spanish | 1436 | 1142 |
Instruction Following GLM-4.6 leads
GLM-4.6: 74.5 (#93), Llama 2-70B: 55.1 (#273)
| Benchmark | GLM-4.6 | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1410 | 1071 |
Long Context GLM-4.6 leads
GLM-4.6: 43.4 (#91), Llama 2-70B: 32.2 (#265)
| Benchmark | GLM-4.6 | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1423 | 1062 |
Writing & Preference GLM-4.6 leads
GLM-4.6: 61.3 (#86), Llama 2-70B: 32.5 (#274)
| Benchmark | GLM-4.6 | Llama 2-70B |
|---|---|---|
| LMArena Text | 1440 | 1115 |
| LMArena Creative Writing | 1411 | 1075 |
| LMArena Multi-Turn | 1427 | 1088 |
| EQ-Bench Creative Writing | 1411 | — |
Frequently asked questions
Is GLM-4.6 better than Llama 2-70B?
GLM-4.6 is the stronger model overall, scoring 40.9 to 24.1 on the Noometry Index.
Is GLM-4.6 or Llama 2-70B better for coding?
GLM-4.6 scores higher on coding benchmarks: 39.9 versus 30.8 in the Noometry coding category.
How many benchmarks do GLM-4.6 and Llama 2-70B share?
17 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Llama 2-70B has 35.