Model comparison
GLM-4.7-Flash vs GPT-5.4 nano
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 38.4 on the Noometry Index. GLM-4.7-Flash costs 3.2× less per token, which makes it the better buy when GPT-5.4 nano's lead doesn't matter for your workload.
Last verified . 20 shared benchmarks.
Summary
- They share 20 benchmarks with published results for both. GLM-4.7-Flash scores higher in 0 categories and GPT-5.4 nano in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-5.4 nano leads 55.9 to 47.7.
- The biggest single-benchmark swing is Chess Puzzles: 0% for GLM-4.7-Flash and 30% for GPT-5.4 nano.
- GLM-4.7-Flash is cheaper at $0.06 / $0.40 per million input/output tokens, against $0.20 / $1.25 for GPT-5.4 nano.
- GPT-5.4 nano accepts more context: 400K tokens versus 200K.
- GLM-4.7-Flash has downloadable open weights; the other is API-only.
Side by side
| GLM-4.7-Flash | GPT-5.4 nano | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 38.4 | 41.9 |
| Released | 2026-01-19 | 2026-03-17 |
| Weights | Open | Proprietary |
| Context window | 200K | 400K |
| Max output | 131K | 128K |
| Input $ / M tokens | $0.06 | $0.20 |
| Output $ / M tokens | $0.40 | $1.25 |
| Results tracked | 21 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
GLM-4.7-Flash: 40.0 (#133), GPT-5.4 nano: 43.1 (#82)
| Benchmark | GLM-4.7-Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Coding | 1383 | 1404 |
| SciCode | — | 46.9% |
| WeirdML | — | 49.2% |
| ALE-Bench | — | 1,005 |
Reasoning GPT-5.4 nano leads
GLM-4.7-Flash: 20.2 (#224), GPT-5.4 nano: 23.8 (#159)
| Benchmark | GLM-4.7-Flash | GPT-5.4 nano |
|---|---|---|
| Chess Puzzles | 0% | 30% |
| LMArena Hard Prompts | 1356 | 1380 |
| ARC-AGI-2 | — | 5.7% |
| Kagi LLM Benchmark | — | 39.7% |
| ARC-AGI-1 | — | 51.5% |
| CritPt | — | 9.3% |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LMCA | — | 36.9% |
| Epoch Capabilities Index | — | 145.78 |
| ForecastBench | — | 57.3 |
Math GPT-5.4 nano leads
GLM-4.7-Flash: 34.4 (#159), GPT-5.4 nano: 40.7 (#76)
| Benchmark | GLM-4.7-Flash | GPT-5.4 nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 58.3% | 87.8% |
| LMArena Math | 1354 | 1406 |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 5% |
| FrontierMath (Feb 2025 set) | — | 25.9% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5.4 nano leads
GLM-4.7-Flash: 35.4 (#177), GPT-5.4 nano: 41.8 (#94)
| Benchmark | GLM-4.7-Flash | GPT-5.4 nano |
|---|---|---|
| GPQA Diamond | 60.5% | 78.5% |
| Vectara Hallucination Rate | 9.3% | 3.1% |
| LMArena Expert | 1357 | 1397 |
| SimpleQA Verified | — | 11.7% |
Multimodal Not comparable
GLM-4.7-Flash: —, GPT-5.4 nano: 36.0 (#73)
| Benchmark | GLM-4.7-Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | — | 1196 |
Multilingual GPT-5.4 nano leads
GLM-4.7-Flash: 46.8 (#152), GPT-5.4 nano: 48.8 (#136)
| Benchmark | GLM-4.7-Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1331 | 1359 |
| LMArena Chinese | 1403 | 1391 |
| LMArena French | 1332 | 1396 |
| LMArena German | 1336 | 1368 |
| LMArena Korean | 1284 | 1321 |
| LMArena Russian | 1333 | 1363 |
| LMArena Spanish | 1351 | 1368 |
| LMArena Japanese | — | 1342 |
Instruction Following GPT-5.4 nano leads
GLM-4.7-Flash: 70.3 (#162), GPT-5.4 nano: 72.1 (#139)
| Benchmark | GLM-4.7-Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1327 | 1362 |
Long Context Too close to call
GLM-4.7-Flash: 40.9 (#143), GPT-5.4 nano: 41.6 (#131)
| Benchmark | GLM-4.7-Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1345 | 1365 |
Writing & Preference GPT-5.4 nano leads
GLM-4.7-Flash: 47.7 (#206), GPT-5.4 nano: 55.9 (#137)
| Benchmark | GLM-4.7-Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1351 | 1372 |
| LMArena Creative Writing | 1297 | 1313 |
| LMArena Multi-Turn | 1342 | 1382 |
| EQ-Bench Creative Writing | 1125 | — |
Frequently asked questions
Is GLM-4.7-Flash better than GPT-5.4 nano?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 38.4 on the Noometry Index. GLM-4.7-Flash costs 3.2× less per token, which makes it the better buy when GPT-5.4 nano's lead doesn't matter for your workload.
Which is cheaper, GLM-4.7-Flash or GPT-5.4 nano?
GLM-4.7-Flash is cheaper. It lists at $0.06 per million input tokens and $0.40 per million output tokens; GPT-5.4 nano lists at $0.20 and $1.25.
Is GLM-4.7-Flash or GPT-5.4 nano better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.1 versus 40.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 nano does, with 400K tokens against 200K.
How many benchmarks do GLM-4.7-Flash and GPT-5.4 nano share?
20 benchmarks have published results for both models. GLM-4.7-Flash has 21 scored results on Noometry and GPT-5.4 nano has 40.