Model comparison
GLM-4.6 vs Llama 3.1-70B
GLM-4.6 is the stronger model overall, scoring 40.9 to 29.4 on the Noometry Index. Llama 3.1-70B costs 2.5× less per token, which makes it the better buy when GLM-4.6's lead doesn't matter for your workload.
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 3.1-70B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-4.6 leads 61.3 to 35.6.
- Llama 3.1-70B is cheaper at $0.40 / $0.40 per million input/output tokens, against $0.60 / $2.20 for GLM-4.6.
- GLM-4.6 accepts more context: 205K tokens versus 128K.
Side by side
| GLM-4.6 | Llama 3.1-70B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 40.9 | 29.4 |
| Released | 2025-09-30 | 2024-07-23 |
| Weights | Open | Open |
| Context window | 205K | 128K |
| Max output | 131K | 4K |
| Input $ / M tokens | $0.60 | $0.40 |
| Output $ / M tokens | $2.20 | $0.40 |
| 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 3.1-70B: 29.8 (#292)
| Benchmark | GLM-4.6 | Llama 3.1-70B |
|---|---|---|
| LMArena Coding | 1449 | 1260 |
| SWE-bench Verified (bash only) | 55.4% | — |
| LMArena WebDev | 1339 | — |
| SciCode | 38.4% | — |
| WeirdML | — | 9% |
| BigCodeBench Instruct | — | 46.1% |
| BigCodeBench Complete | — | 54.8% |
| ALE-Bench | 340.82 | — |
Agentic & Tool Use GLM-4.6 leads
GLM-4.6: 32.1 (#60), Llama 3.1-70B: 24.9 (#104)
| Benchmark | GLM-4.6 | Llama 3.1-70B |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |
| TheAgentCompany | — | 6.9% |
| BALROG | — | 27.9% |
Reasoning GLM-4.6 leads
GLM-4.6: 23.1 (#168), Llama 3.1-70B: 20.8 (#215)
| Benchmark | GLM-4.6 | Llama 3.1-70B |
|---|---|---|
| LMArena Hard Prompts | 1440 | 1241 |
| Kagi LLM Benchmark | 47.4% | — |
| CritPt | 1.1% | — |
| DTBench | — | 60% |
| LMCA | — | 14.8% |
| Epoch Capabilities Index | — | 125.91 |
Math GLM-4.6 leads
GLM-4.6: 36.4 (#115), Llama 3.1-70B: 12.7 (#298)
| Benchmark | GLM-4.6 | Llama 3.1-70B |
|---|---|---|
| LMArena Math | 1431 | 1252 |
| OTIS Mock AIME 2024-2025 | — | 3.6% |
| Omni-MATH | — | 21% |
| MATH Level 5 | — | 36.7% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-4.6 leads
GLM-4.6: 39.6 (#118), Llama 3.1-70B: 24.1 (#264)
| Benchmark | GLM-4.6 | Llama 3.1-70B |
|---|---|---|
| LMArena Expert | 1432 | 1209 |
| GPQA Diamond | — | 44.2% |
| MMLU-Pro | — | 65.3% |
| Vectara Hallucination Rate | 9.5% | — |
| GPQA (HELM) | — | 42.6% |
| MMLU | — | 80.1% |
Multilingual GLM-4.6 leads
GLM-4.6: 53.6 (#62), Llama 3.1-70B: 38.9 (#220)
| Benchmark | GLM-4.6 | Llama 3.1-70B |
|---|---|---|
| LMArena Non-English | 1426 | 1219 |
| LMArena Chinese | 1500 | 1215 |
| LMArena French | 1459 | 1260 |
| LMArena German | 1447 | 1222 |
| LMArena Japanese | 1394 | 1132 |
| LMArena Korean | 1401 | 1140 |
| LMArena Russian | 1418 | 1234 |
| LMArena Spanish | 1436 | 1253 |
Instruction Following GLM-4.6 leads
GLM-4.6: 74.5 (#93), Llama 3.1-70B: 65.5 (#218)
| Benchmark | GLM-4.6 | Llama 3.1-70B |
|---|---|---|
| LMArena Instruction Following | 1410 | 1231 |
| IFEval | — | 82.1% |
Long Context GLM-4.6 leads
GLM-4.6: 43.4 (#91), Llama 3.1-70B: 37.6 (#208)
| Benchmark | GLM-4.6 | Llama 3.1-70B |
|---|---|---|
| LMArena Longer Query | 1423 | 1241 |
Writing & Preference GLM-4.6 leads
GLM-4.6: 61.3 (#86), Llama 3.1-70B: 35.6 (#262)
| Benchmark | GLM-4.6 | Llama 3.1-70B |
|---|---|---|
| LMArena Text | 1440 | 1261 |
| LMArena Creative Writing | 1411 | 1232 |
| EQ-Bench Creative Writing | 1411 | 784 |
| LMArena Multi-Turn | 1427 | 1256 |
| WildBench | — | 75.8% |
Frequently asked questions
Is GLM-4.6 better than Llama 3.1-70B?
GLM-4.6 is the stronger model overall, scoring 40.9 to 29.4 on the Noometry Index. Llama 3.1-70B costs 2.5× less per token, which makes it the better buy when GLM-4.6's lead doesn't matter for your workload.
Which is cheaper, GLM-4.6 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-4.6 lists at $0.60 and $2.20.
Is GLM-4.6 or Llama 3.1-70B better for coding?
GLM-4.6 scores higher on coding benchmarks: 39.9 versus 29.8 in the Noometry coding category.
Which has the bigger context window?
GLM-4.6 does, with 205K tokens against 128K.
How many benchmarks do GLM-4.6 and Llama 3.1-70B share?
18 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Llama 3.1-70B has 35.