Model comparison
GLM-5.3 vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 54.8 on the Noometry Index. GLM-5.3 costs 2.8× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 40 shared benchmarks.
Summary
- They share 40 benchmarks with published results for both. GLM-5.3 scores higher in 0 categories and Kimi K3 in 9 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 46.1.
- The biggest single-benchmark swing is ProofBench: 49% for GLM-5.3 and 87% for Kimi K3.
- GLM-5.3 is cheaper at $1.40 / $4.40 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 1M.
Side by side
| GLM-5.3 | Kimi K3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Moonshot AI |
| Noometry Index | 54.8 | 59.5 |
| Released | 2026-08-14 | 2026-07-16 |
| Weights | Open | Open |
| Context window | 1M | 1.05M |
| Max output | 131K | 1.05M |
| Input $ / M tokens | $1.40 | $3 |
| Output $ / M tokens | $4.40 | $15 |
| Results tracked | 42 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
GLM-5.3: 59.5 (#14), Kimi K3: 61.0 (#10)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| DeepSWE | 69% | 68.5% |
| FrontierCode | 40.1% | 44.2% |
| LMArena WebDev | 1622 | 1654 |
| FrontierSWE | 30.2% | 25.9% |
| SciCode | 59% | 59.5% |
| WeirdML | 75.4% | 82.6% |
| LMArena Coding | 1496 | 1508 |
| ALE-Bench | 1,317 | 1,524 |
| CursorBench | 42.6% | — |
Agentic & Tool Use Kimi K3 leads
GLM-5.3: 36.4 (#38), Kimi K3: 41.8 (#20)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| APEX-Agents | 56.6% | 50.6% |
| Vending-Bench 2 | 8,164 | 5,165 |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
Reasoning Kimi K3 leads
GLM-5.3: 46.1 (#46), Kimi K3: 63.0 (#17)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| NYT Connections (extended) | 74.2% | 93.6% |
| CritPt | 19.1% | 23.4% |
| Chess Puzzles | 21% | 39% |
| LMArena Hard Prompts | 1489 | 1496 |
| Mystery Game Puzzles | 33% | 26% |
| DTBench | 87.7% | 91.2% |
| LMCA | 55.5% | 52.7% |
| Epoch Capabilities Index | 155.61 | 157.45 |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | — | 60.7% |
| ARC-AGI-1 | — | 94.5% |
| Surface Evolver Bench | — | 95% |
| Bench to the Future 3 | 0.15 | — |
| ForecastBench | — | 61.1 |
Math Kimi K3 leads
GLM-5.3: 62.3 (#33), Kimi K3: 74.2 (#16)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 68.8% | 72.2% |
| FrontierMath Tier 4 | 29.3% | 39% |
| OTIS Mock AIME 2024-2025 | 91.1% | 97.2% |
| ProofBench | 49% | 87% |
| LMArena Math | 1489 | 1491 |
| MathArena Final-Answer Competitions | — | 87.8% |
Knowledge Kimi K3 leads
GLM-5.3: 58.3 (#37), Kimi K3: 63.2 (#21)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| GPQA Diamond | 90.9% | 93.1% |
| SimpleQA Verified | 41% | 50.6% |
| LMArena Expert | 1516 | 1521 |
Multimodal Not comparable
GLM-5.3: —, Kimi K3: 37.8 (#70)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Too close to call
GLM-5.3: 55.7 (#28), Kimi K3: 56.3 (#21)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1457 | 1466 |
| LMArena Chinese | 1528 | 1529 |
| LMArena French | 1499 | 1491 |
| LMArena German | 1499 | 1488 |
| LMArena Japanese | 1453 | 1487 |
| LMArena Korean | 1472 | 1458 |
| LMArena Russian | 1463 | 1482 |
| LMArena Spanish | 1460 | 1472 |
Instruction Following Too close to call
GLM-5.3: 77.5 (#23), Kimi K3: 77.7 (#14)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1477 | 1483 |
Long Context Too close to call
GLM-5.3: 45.4 (#41), Kimi K3: 45.8 (#29)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1482 | 1494 |
Writing & Preference Too close to call
GLM-5.3: 75.7 (#6), Kimi K3: 76.6 (#4)
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
| LMArena Text | 1471 | 1476 |
| LMArena Creative Writing | 1457 | 1454 |
| EQ-Bench Creative Writing | 2075 | 2082 |
| LMArena Multi-Turn | 1472 | 1488 |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is GLM-5.3 better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 54.8 on the Noometry Index. GLM-5.3 costs 2.8× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, GLM-5.3 or Kimi K3?
GLM-5.3 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; Kimi K3 lists at $3 and $15.
Is GLM-5.3 or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 59.5 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 1M.
How many benchmarks do GLM-5.3 and Kimi K3 share?
40 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Kimi K3 has 53.