Model comparison
GLM-4.7-Flash vs Llama-3.3-70B-Instruct
GLM-4.7-Flash is the stronger model overall, scoring 38.4 to 30.3 on the Noometry Index.
Last verified . 19 shared benchmarks.
MetaSummary
- They share 19 benchmarks with published results for both. GLM-4.7-Flash scores higher in 6 categories and Llama-3.3-70B-Instruct in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-4.7-Flash leads 34.4 to 14.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 58.3% for GLM-4.7-Flash and 5.1% for Llama-3.3-70B-Instruct.
- GLM-4.7-Flash is cheaper at $0.06 / $0.40 per million input/output tokens, against $0.10 / $0.32 for Llama-3.3-70B-Instruct.
- GLM-4.7-Flash accepts more context: 200K tokens versus 128K.
Side by side
| GLM-4.7-Flash | Llama-3.3-70B-Instruct | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Meta |
| Noometry Index | 38.4 | 30.3 |
| Released | 2026-01-19 | 2024-12-06 |
| Weights | Open | Open |
| Context window | 200K | 128K |
| Max output | 131K | 4K |
| Input $ / M tokens | $0.06 | $0.10 |
| Output $ / M tokens | $0.40 | $0.32 |
| Results tracked | 21 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-4.7-Flash leads
GLM-4.7-Flash: 40.0 (#133), Llama-3.3-70B-Instruct: 30.6 (#284)
| Benchmark | GLM-4.7-Flash | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Coding | 1383 | 1268 |
| SciCode | — | 26% |
| WeirdML | — | 14.4% |
| BigCodeBench Instruct | — | 46.9% |
| LiveBench Coding | — | 36.6% |
| BigCodeBench Complete | — | 57.5% |
Agentic & Tool Use Not comparable
GLM-4.7-Flash: —, Llama-3.3-70B-Instruct: 25.6 (#100)
| Benchmark | GLM-4.7-Flash | Llama-3.3-70B-Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 31.9% |
| BALROG | — | 23% |
Reasoning GLM-4.7-Flash leads
GLM-4.7-Flash: 20.2 (#224), Llama-3.3-70B-Instruct: 13.7 (#325)
| Benchmark | GLM-4.7-Flash | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Hard Prompts | 1356 | 1257 |
| SimpleBench | — | 19.9% |
| CritPt | — | 0% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 50.8% |
| DTBench | — | 59.5% |
| LiveBench Data Analysis | — | 49.5% |
| LMCA | — | 17.5% |
| Epoch Capabilities Index | — | 127.32 |
| ForecastBench | — | 58.6 |
| LiveBench | — | 50.2% |
Math GLM-4.7-Flash leads
GLM-4.7-Flash: 34.4 (#159), Llama-3.3-70B-Instruct: 14.3 (#292)
| Benchmark | GLM-4.7-Flash | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 58.3% | 5.1% |
| LMArena Math | 1354 | 1268 |
| LiveBench Math | — | 42.2% |
| MATH Level 5 | — | 41.6% |
Knowledge GLM-4.7-Flash leads
GLM-4.7-Flash: 35.4 (#177), Llama-3.3-70B-Instruct: 30.5 (#220)
| Benchmark | GLM-4.7-Flash | Llama-3.3-70B-Instruct |
|---|---|---|
| GPQA Diamond | 60.5% | 47.4% |
| Vectara Hallucination Rate | 9.3% | 4.1% |
| LMArena Expert | 1357 | 1225 |
| Confabulations | — | 22.8% |
| MMLU | — | 86.3% |
Multilingual GLM-4.7-Flash leads
GLM-4.7-Flash: 46.8 (#152), Llama-3.3-70B-Instruct: 40.1 (#215)
| Benchmark | GLM-4.7-Flash | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1331 | 1236 |
| LMArena Chinese | 1403 | 1217 |
| LMArena French | 1332 | 1280 |
| LMArena German | 1336 | 1251 |
| LMArena Korean | 1284 | 1142 |
| LMArena Russian | 1333 | 1252 |
| LMArena Spanish | 1351 | 1269 |
| LMArena Japanese | — | 1149 |
Instruction Following Too close to call
GLM-4.7-Flash: 70.3 (#162), Llama-3.3-70B-Instruct: 71.3 (#152)
| Benchmark | GLM-4.7-Flash | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Instruction Following | 1327 | 1242 |
| LiveBench Instruction Following | — | 82.7% |
Long Context GLM-4.7-Flash leads
GLM-4.7-Flash: 40.9 (#143), Llama-3.3-70B-Instruct: 26.4 (#290)
| Benchmark | GLM-4.7-Flash | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1345 | 1256 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference Too close to call
GLM-4.7-Flash: 47.7 (#206), Llama-3.3-70B-Instruct: 47.8 (#203)
| Benchmark | GLM-4.7-Flash | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1351 | 1274 |
| LMArena Creative Writing | 1297 | 1250 |
| LMArena Multi-Turn | 1342 | 1279 |
| EQ-Bench Creative Writing | 1125 | — |
| LiveBench Language | — | 39.2% |
Frequently asked questions
Is GLM-4.7-Flash better than Llama-3.3-70B-Instruct?
GLM-4.7-Flash is the stronger model overall, scoring 38.4 to 30.3 on the Noometry Index.
Which is cheaper, GLM-4.7-Flash or Llama-3.3-70B-Instruct?
GLM-4.7-Flash is cheaper. It lists at $0.06 per million input tokens and $0.40 per million output tokens; Llama-3.3-70B-Instruct lists at $0.10 and $0.32.
Is GLM-4.7-Flash or Llama-3.3-70B-Instruct better for coding?
GLM-4.7-Flash scores higher on coding benchmarks: 40.0 versus 30.6 in the Noometry coding category.
Which has the bigger context window?
GLM-4.7-Flash does, with 200K tokens against 128K.
How many benchmarks do GLM-4.7-Flash and Llama-3.3-70B-Instruct share?
19 benchmarks have published results for both models. GLM-4.7-Flash has 21 scored results on Noometry and Llama-3.3-70B-Instruct has 43.