Model comparison
Command A vs GLM-5.3-Flash
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 36.5 on the Noometry Index.
Last verified . 17 shared benchmarks.
CohereSummary
- They share 17 benchmarks with published results for both. Command A scores higher in 1 category and GLM-5.3-Flash in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.3-Flash leads 48.0 to 18.3.
- GLM-5.3-Flash is cheaper at $0.15 / $0.50 per million input/output tokens, against $2.50 / $10 for Command A.
- GLM-5.3-Flash accepts more context: 1M tokens versus 256K.
Side by side
| Command A | GLM-5.3-Flash | |
|---|---|---|
| Provider | Cohere | Z.ai (Zhipu) |
| Noometry Index | 36.5 | 51.8 |
| Released | 2025-03-13 | 2026-08-20 |
| Weights | Open | Open |
| Context window | 256K | 1M |
| Max output | 8K | 131K |
| Input $ / M tokens | $2.50 | $0.15 |
| Output $ / M tokens | $10 | $0.50 |
| Results tracked | 24 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3-Flash leads
Command A: 27.2 (#322), GLM-5.3-Flash: 53.1 (#31)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| LMArena Coding | 1330 | 1508 |
| DeepSWE | — | 63.4% |
| FrontierCode | — | 31.8% |
| Aider Polyglot | 12% | — |
| CursorBench | — | 36.8% |
| LMArena WebDev | — | 1609 |
| FrontierSWE | — | 18.1% |
| SciCode | — | 51.6% |
| ALE-Bench | — | 303.55 |
Agentic & Tool Use Command A leads
Command A: 35.9 (#40), GLM-5.3-Flash: 34.2 (#47)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| APEX-Agents | — | 52.8% |
| Berkeley Function Calling Leaderboard | 57.1% | — |
| GDP.pdf | — | 14% |
Reasoning GLM-5.3-Flash leads
Command A: 18.3 (#283), GLM-5.3-Flash: 48.0 (#42)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| LMArena Hard Prompts | 1326 | 1491 |
| ARC-AGI-2 | — | 65.8% |
| Kagi LLM Benchmark | 28.8% | — |
| ARC-AGI-1 | — | 91% |
| CritPt | — | 15.4% |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| DTBench | 61.3% | — |
| LMCA | 10.3% | — |
| Surface Evolver Bench | — | 52.5% |
| Bench to the Future 3 | — | 0.15 |
| Epoch Capabilities Index | — | 151.88 |
Math GLM-5.3-Flash leads
Command A: 36.2 (#171), GLM-5.3-Flash: 53.3 (#47)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| LMArena Math | 1300 | 1500 |
| FrontierMath (Tiers 1-3) | — | 55.8% |
| FrontierMath Tier 4 | — | 17.1% |
| OTIS Mock AIME 2024-2025 | — | 93.9% |
| ProofBench | — | 21% |
Knowledge GLM-5.3-Flash leads
Command A: 37.1 (#159), GLM-5.3-Flash: 58.4 (#36)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| LMArena Expert | 1295 | 1513 |
| GPQA Diamond | — | 90.2% |
| Vectara Hallucination Rate | 9.3% | — |
Multimodal Not comparable
Command A: —, GLM-5.3-Flash: 42.8 (#27)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| LMArena Vision | — | 1296 |
Multilingual GLM-5.3-Flash leads
Command A: 45.3 (#170), GLM-5.3-Flash: 56.0 (#25)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| LMArena Non-English | 1313 | 1462 |
| LMArena Chinese | 1327 | 1527 |
| LMArena French | 1351 | 1496 |
| LMArena German | 1341 | 1470 |
| LMArena Japanese | 1285 | 1429 |
| LMArena Korean | 1285 | 1446 |
| LMArena Russian | 1314 | 1469 |
| LMArena Spanish | 1347 | 1471 |
Instruction Following GLM-5.3-Flash leads
Command A: 69.1 (#177), GLM-5.3-Flash: 77.5 (#20)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| LMArena Instruction Following | 1309 | 1478 |
Long Context GLM-5.3-Flash leads
Command A: 40.6 (#151), GLM-5.3-Flash: 45.4 (#39)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| LMArena Longer Query | 1334 | 1482 |
Writing & Preference GLM-5.3-Flash leads
Command A: 47.6 (#208), GLM-5.3-Flash: 65.3 (#50)
| Benchmark | Command A | GLM-5.3-Flash |
|---|---|---|
| LMArena Text | 1331 | 1471 |
| LMArena Creative Writing | 1319 | 1442 |
| LMArena Multi-Turn | 1339 | 1467 |
| EQ-Bench Creative Writing | 1145 | — |
Frequently asked questions
Is Command A better than GLM-5.3-Flash?
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 36.5 on the Noometry Index.
Which is cheaper, Command A or GLM-5.3-Flash?
GLM-5.3-Flash is cheaper. It lists at $0.15 per million input tokens and $0.50 per million output tokens; Command A lists at $2.50 and $10.
Is Command A or GLM-5.3-Flash better for coding?
GLM-5.3-Flash scores higher on coding benchmarks: 53.1 versus 27.2 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3-Flash does, with 1M tokens against 256K.
How many benchmarks do Command A and GLM-5.3-Flash share?
17 benchmarks have published results for both models. Command A has 24 scored results on Noometry and GLM-5.3-Flash has 40.