Model comparison
GLM-4.7-Flash vs GPT-6 Astra
GPT-6 Astra is the stronger model overall, scoring 70.8 to 38.8 on the Noometry Index. GLM-4.7-Flash costs 138× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
Summary
- They share 21 benchmarks with published results for both. GLM-4.7-Flash scores higher in 0 categories and GPT-6 Astra in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Astra leads 85.1 to 20.9.
- The biggest single-benchmark swing is Chess Puzzles: 0% for GLM-4.7-Flash and 72% for GPT-6 Astra.
- GLM-4.7-Flash is cheaper at $0.06 / $0.40 per million input/output tokens, against $10 / $50 for GPT-6 Astra.
- GPT-6 Astra accepts more context: 1.05M tokens versus 200K.
- GLM-4.7-Flash has downloadable open weights; the other is API-only.
Side by side
| GLM-4.7-Flash | GPT-6 Astra | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 38.8 | 70.8 |
| Released | 2026-01-19 | 2026-09-03 |
| Weights | Open | Proprietary |
| Context window | 200K | 1.05M |
| Max output | 131K | 128K |
| Input $ / M tokens | $0.06 | $10 |
| Output $ / M tokens | $0.40 | $50 |
| Results tracked | 21 | 56 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Astra leads
GLM-4.7-Flash: 40.6 (#135), GPT-6 Astra: 73.7 (#2)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| LMArena Coding | 1383 | 1487 |
| DeepSWE | — | 74.1% |
| FrontierCode | — | 53.3% |
| LMArena WebDev | — | 1786 |
| FrontierSWE | — | 65.5% |
| SciCode | — | 56.5% |
| GSO | — | 79.4% |
| WeirdML | — | 93.6% |
| MirrorCode | — | 46.7% |
| ALE-Bench | — | 2,951 |
Agentic & Tool Use Not comparable
GLM-4.7-Flash: —, GPT-6 Astra: 52.9 (#3)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| APEX-Agents | — | 64.7% |
| Remote Labor Index | — | 20.8% |
| BALROG | — | 68.3% |
| GDP.pdf | — | 34.2% |
| Vending-Bench 2 | — | 15,515 |
Reasoning GPT-6 Astra leads
GLM-4.7-Flash: 20.9 (#229), GPT-6 Astra: 85.1 (#1)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| Chess Puzzles | 0% | 72% |
| LMArena Hard Prompts | 1356 | 1462 |
| ARC-AGI-2 | — | 95% |
| NYT Connections (extended) | — | 98.1% |
| ARC-AGI-1 | — | 98.5% |
| CritPt | — | 31.7% |
| EBR-Bench | — | 76.2% |
| Mystery Game Puzzles | — | 84% |
| DTBench | — | 97.3% |
| LMCA | — | 64.4% |
| Bench to the Future 3 | — | 0.14 |
| Epoch Capabilities Index | — | 166.45 |
Math GPT-6 Astra leads
GLM-4.7-Flash: 36.1 (#173), GPT-6 Astra: 93.5 (#2)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 58.3% | 100% |
| LMArena Math | 1355 | 1465 |
| FrontierMath (Tiers 1-3) | — | 93.7% |
| FrontierMath Tier 4 | — | 97.6% |
| ProofBench | — | 99% |
| FrontierMath Erdős | — | 2.9% |
Knowledge GPT-6 Astra leads
GLM-4.7-Flash: 35.5 (#184), GPT-6 Astra: 75.3 (#1)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| GPQA Diamond | 60.5% | 95.8% |
| Vectara Hallucination Rate | 9.3% | 8.7% |
| LMArena Expert | 1357 | 1483 |
| Humanity's Last Exam | — | 54.8% |
| SimpleQA Verified | — | 75.6% |
Multimodal Not comparable
GLM-4.7-Flash: —, GPT-6 Astra: 55.0 (#3)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| LMArena Vision | — | 1281 |
| Blueprint-Bench 2 | — | 49.7% |
| Furniture Assembly | — | 80% |
| LMArena Document | — | 1468 |
Multilingual GPT-6 Astra leads
GLM-4.7-Flash: 46.5 (#158), GPT-6 Astra: 53.7 (#61)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| LMArena Non-English | 1330 | 1430 |
| LMArena Chinese | 1403 | 1484 |
| LMArena French | 1332 | 1456 |
| LMArena German | 1337 | 1440 |
| LMArena Korean | 1283 | 1426 |
| LMArena Russian | 1332 | 1436 |
| LMArena Spanish | 1350 | 1407 |
| LMArena Japanese | — | 1379 |
Instruction Following GPT-6 Astra leads
GLM-4.7-Flash: 70.1 (#167), GPT-6 Astra: 76.3 (#44)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| LMArena Instruction Following | 1327 | 1450 |
Long Context GPT-6 Astra leads
GLM-4.7-Flash: 40.9 (#148), GPT-6 Astra: 44.5 (#62)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| LMArena Longer Query | 1345 | 1456 |
Writing & Preference GPT-6 Astra leads
GLM-4.7-Flash: 47.4 (#210), GPT-6 Astra: 75.3 (#7)
| Benchmark | GLM-4.7-Flash | GPT-6 Astra |
|---|---|---|
| LMArena Text | 1351 | 1441 |
| LMArena Creative Writing | 1297 | 1418 |
| EQ-Bench Creative Writing | 1125 | 2173 |
| LMArena Multi-Turn | 1342 | 1448 |
Frequently asked questions
Is GLM-4.7-Flash better than GPT-6 Astra?
GPT-6 Astra is the stronger model overall, scoring 70.8 to 38.8 on the Noometry Index. GLM-4.7-Flash costs 138× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Which is cheaper, GLM-4.7-Flash or GPT-6 Astra?
GLM-4.7-Flash is cheaper. It lists at $0.06 per million input tokens and $0.40 per million output tokens; GPT-6 Astra lists at $10 and $50.
Is GLM-4.7-Flash or GPT-6 Astra better for coding?
GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 40.6 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Astra does, with 1.05M tokens against 200K.
How many benchmarks do GLM-4.7-Flash and GPT-6 Astra share?
21 benchmarks have published results for both models. GLM-4.7-Flash has 21 scored results on Noometry and GPT-6 Astra has 56.