Model comparison
GLM-4.6 vs Llama 2-7B
GLM-4.6 is the stronger model overall, scoring 40.9 to 28.6 on the Noometry Index.
Last verified . 15 shared benchmarks.
MetaSummary
- They share 15 benchmarks with published results for both. GLM-4.6 scores higher in 8 categories and Llama 2-7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-4.6 leads 61.3 to 28.1.
Side by side
| GLM-4.6 | Llama 2-7B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 40.9 | 28.6 |
| 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 | 29 |
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 2-7B: 28.7 (#302)
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| LMArena Coding | 1449 | 1002 |
| 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 2-7B: —
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |
Reasoning GLM-4.6 leads
GLM-4.6: 23.1 (#168), Llama 2-7B: 15.1 (#309)
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| LMArena Hard Prompts | 1440 | 1009 |
| Kagi LLM Benchmark | 47.4% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 0% |
| BIG-Bench Hard | — | 39.2% |
| Epoch Capabilities Index | — | 98.54 |
| HellaSwag | — | 77.2% |
| LAMBADA | — | 73.3% |
| PIQA | — | 78.8% |
| WinoGrande | — | 69.2% |
Math GLM-4.6 leads
GLM-4.6: 36.4 (#115), Llama 2-7B: 28.2 (#231)
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| LMArena Math | 1431 | 1042 |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
| GSM8K | — | 16.7% |
Knowledge GLM-4.6 leads
GLM-4.6: 39.6 (#118), Llama 2-7B: 27.7 (#243)
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| LMArena Expert | 1432 | 1036 |
| Vectara Hallucination Rate | 9.5% | — |
| ARC (AI2) Challenge | — | 45.9% |
| BoolQ | — | 77.9% |
| MMLU | — | 45.8% |
| OpenBookQA | — | 58.6% |
| TriviaQA | — | 73.7% |
Multimodal Not comparable
GLM-4.6: —, Llama 2-7B: —
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| ScienceQA | — | 43.1% |
Multilingual GLM-4.6 leads
GLM-4.6: 53.6 (#62), Llama 2-7B: 23.9 (#288)
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| LMArena Non-English | 1426 | 973 |
| LMArena Chinese | 1500 | 973 |
| LMArena French | 1459 | 969 |
| LMArena German | 1447 | 978 |
| LMArena Russian | 1418 | 995 |
| LMArena Spanish | 1436 | 1007 |
| LMArena Japanese | 1394 | — |
| LMArena Korean | 1401 | — |
Instruction Following GLM-4.6 leads
GLM-4.6: 74.5 (#93), Llama 2-7B: 51.0 (#293)
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| LMArena Instruction Following | 1410 | 1006 |
Long Context GLM-4.6 leads
GLM-4.6: 43.4 (#91), Llama 2-7B: 30.4 (#282)
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| LMArena Longer Query | 1423 | 999 |
Writing & Preference GLM-4.6 leads
GLM-4.6: 61.3 (#86), Llama 2-7B: 28.1 (#293)
| Benchmark | GLM-4.6 | Llama 2-7B |
|---|---|---|
| LMArena Text | 1440 | 1053 |
| LMArena Creative Writing | 1411 | 1033 |
| LMArena Multi-Turn | 1427 | 1029 |
| EQ-Bench Creative Writing | 1411 | — |
Frequently asked questions
Is GLM-4.6 better than Llama 2-7B?
GLM-4.6 is the stronger model overall, scoring 40.9 to 28.6 on the Noometry Index.
Is GLM-4.6 or Llama 2-7B better for coding?
GLM-4.6 scores higher on coding benchmarks: 39.9 versus 28.7 in the Noometry coding category.
How many benchmarks do GLM-4.6 and Llama 2-7B share?
15 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Llama 2-7B has 29.