Model comparison
GLM-4.6V vs Llama-3.3-70B-Instruct
GLM-4.6V is the stronger model overall, scoring 41.3 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 2.9× less per token, which makes it the better buy when GLM-4.6V's lead doesn't matter for your workload.
Last verified . 11 shared benchmarks.
MetaSummary
- They share 11 benchmarks with published results for both. GLM-4.6V scores higher in 7 categories and Llama-3.3-70B-Instruct in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in long context, where GLM-4.6V leads 41.3 to 26.4.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.30 / $0.90 for GLM-4.6V.
Side by side
| GLM-4.6V | Llama-3.3-70B-Instruct | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 41.3 | 30.6 |
| Released | 2025-12-08 | 2024-12-06 |
| Weights | Open | Open |
| Context window | 128K | 128K |
| Max output | 33K | 4K |
| Input $ / M tokens | $0.30 | $0.10 |
| Output $ / M tokens | $0.90 | $0.32 |
| Results tracked | 12 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-4.6V leads
GLM-4.6V: 40.9 (#128), Llama-3.3-70B-Instruct: 31.0 (#290)
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Coding | 1390 | 1268 |
| SciCode | — | 26% |
| WeirdML | — | 14.4% |
| BigCodeBench Instruct | — | 46.9% |
| LiveBench Coding | — | 36.6% |
| BigCodeBench Complete | — | 57.5% |
Agentic & Tool Use Not comparable
GLM-4.6V: —, Llama-3.3-70B-Instruct: 25.8 (#105)
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 31.9% |
| BALROG | — | 23% |
Reasoning GLM-4.6V leads
GLM-4.6V: 27.6 (#115), Llama-3.3-70B-Instruct: 14.1 (#327)
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Hard Prompts | 1368 | 1257 |
| SimpleBench | — | 19.9% |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 50.8% |
| DTBench | — | 59.5% |
| LiveBench Data Analysis | — | 49.5% |
| LMCA | — | 17.5% |
| Epoch Capabilities Index | — | 127.33 |
| ForecastBench | — | 58.6 |
| LiveBench | — | 50.2% |
Math Not comparable
GLM-4.6V: —, Llama-3.3-70B-Instruct: 15.3 (#298)
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 5.1% |
| LiveBench Math | — | 42.2% |
| LMArena Math | — | 1267 |
| MATH Level 5 | — | 41.6% |
Knowledge GLM-4.6V leads
GLM-4.6V: 38.0 (#149), Llama-3.3-70B-Instruct: 30.6 (#226)
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Expert | 1371 | 1225 |
| GPQA Diamond | — | 47.4% |
| Confabulations | — | 22.8% |
| Vectara Hallucination Rate | — | 4.1% |
| MMLU | — | 86.3% |
Multimodal Not comparable
GLM-4.6V: 34.8 (#90), Llama-3.3-70B-Instruct: —
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Vision | 1164 | — |
Multilingual GLM-4.6V leads
GLM-4.6V: 48.6 (#141), Llama-3.3-70B-Instruct: 39.9 (#220)
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1359 | 1236 |
| LMArena Chinese | 1425 | 1217 |
| LMArena Russian | 1340 | 1252 |
| LMArena French | — | 1281 |
| LMArena German | — | 1251 |
| LMArena Japanese | — | 1150 |
| LMArena Korean | — | 1143 |
| LMArena Spanish | — | 1270 |
Instruction Following Too close to call
GLM-4.6V: 71.4 (#151), Llama-3.3-70B-Instruct: 71.1 (#157)
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Instruction Following | 1352 | 1242 |
| LiveBench Instruction Following | — | 82.7% |
Long Context GLM-4.6V leads
GLM-4.6V: 41.3 (#143), Llama-3.3-70B-Instruct: 26.4 (#295)
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1358 | 1256 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference GLM-4.6V leads
GLM-4.6V: 56.6 (#137), Llama-3.3-70B-Instruct: 47.6 (#207)
| Benchmark | GLM-4.6V | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1377 | 1274 |
| LMArena Creative Writing | 1347 | 1250 |
| LMArena Multi-Turn | 1360 | 1280 |
| LiveBench Language | — | 39.2% |
Frequently asked questions
Is GLM-4.6V better than Llama-3.3-70B-Instruct?
GLM-4.6V is the stronger model overall, scoring 41.3 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 2.9× less per token, which makes it the better buy when GLM-4.6V's lead doesn't matter for your workload.
Which is cheaper, GLM-4.6V or Llama-3.3-70B-Instruct?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; GLM-4.6V lists at $0.30 and $0.90.
Is GLM-4.6V or Llama-3.3-70B-Instruct better for coding?
GLM-4.6V scores higher on coding benchmarks: 40.9 versus 31.0 in the Noometry coding category.
Which has the bigger context window?
Both accept 128K tokens.
How many benchmarks do GLM-4.6V and Llama-3.3-70B-Instruct share?
11 benchmarks have published results for both models. GLM-4.6V has 12 scored results on Noometry and Llama-3.3-70B-Instruct has 43.