Model comparison
GLM-5.1 vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 47.8 on the Noometry Index. GLM-5.1 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 . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. GLM-5.1 scores higher in 0 categories and Kimi K3 in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 49.7.
- The biggest single-benchmark swing is ProofBench: 22.2% for GLM-5.1 and 87% for Kimi K3.
- GLM-5.1 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 200K.
Side by side
| GLM-5.1 | Kimi K3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Moonshot AI |
| Noometry Index | 47.8 | 59.5 |
| Released | 2026-04-07 | 2026-07-16 |
| Weights | Open | Open |
| Context window | 200K | 1.05M |
| Max output | 131K | 1.05M |
| Input $ / M tokens | $1.40 | $3 |
| Output $ / M tokens | $4.40 | $15 |
| Results tracked | 41 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
GLM-5.1: 48.7 (#55), Kimi K3: 61.0 (#10)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| LMArena WebDev | 1508 | 1654 |
| SciCode | 43.8% | 59.5% |
| WeirdML | 57.1% | 82.6% |
| LMArena Coding | 1485 | 1508 |
| ALE-Bench | 887.1 | 1,524 |
| SWE-bench Verified | 74.2% | — |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| FrontierSWE | — | 25.9% |
Agentic & Tool Use Kimi K3 leads
GLM-5.1: 24.9 (#113), Kimi K3: 41.8 (#20)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| APEX-Agents | 40.9% | 50.6% |
| GBAEval | 0% | 48.3% |
| Vending-Bench 2 | 5,634 | 5,165 |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| ExploitBench | 18.1% | — |
| GDP.pdf | — | 19% |
Reasoning Kimi K3 leads
GLM-5.1: 39.1 (#60), Kimi K3: 63.0 (#17)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| SimpleBench | 55.1% | 60.7% |
| NYT Connections (extended) | 77.7% | 93.6% |
| CritPt | 4.6% | 23.4% |
| Chess Puzzles | 19% | 39% |
| LMArena Hard Prompts | 1472 | 1496 |
| Epoch Capabilities Index | 149.84 | 157.45 |
| ARC-AGI-2 | — | 60.4% |
| ARC-AGI-1 | — | 94.5% |
| Thematic Generalization | 69.8% | — |
| Mystery Game Puzzles | — | 26% |
| DTBench | — | 91.2% |
| LMCA | — | 52.7% |
| Surface Evolver Bench | — | 95% |
| ForecastBench | — | 61.1 |
Math Kimi K3 leads
GLM-5.1: 49.7 (#60), Kimi K3: 74.2 (#16)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 36.8% | 72.2% |
| MathArena Final-Answer Competitions | 67.1% | 87.8% |
| OTIS Mock AIME 2024-2025 | 93.3% | 97.2% |
| ProofBench | 22.2% | 87% |
| LMArena Math | 1473 | 1491 |
| FrontierMath Tier 4 | — | 39% |
| FrontierMath (Feb 2025 set) | 33.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge Kimi K3 leads
GLM-5.1: 54.9 (#50), Kimi K3: 63.2 (#21)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| GPQA Diamond | 89.9% | 93.1% |
| SimpleQA Verified | 34% | 50.6% |
| LMArena Expert | 1476 | 1521 |
Multimodal Not comparable
GLM-5.1: —, Kimi K3: 37.8 (#70)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Kimi K3 leads
GLM-5.1: 55.0 (#36), Kimi K3: 56.3 (#21)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1447 | 1466 |
| LMArena Chinese | 1515 | 1529 |
| LMArena French | 1474 | 1491 |
| LMArena German | 1465 | 1488 |
| LMArena Japanese | 1434 | 1487 |
| LMArena Korean | 1418 | 1458 |
| LMArena Russian | 1454 | 1482 |
| LMArena Spanish | 1469 | 1472 |
Instruction Following Kimi K3 leads
GLM-5.1: 76.3 (#42), Kimi K3: 77.7 (#14)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1451 | 1483 |
Long Context Too close to call
GLM-5.1: 44.9 (#53), Kimi K3: 45.8 (#29)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1466 | 1494 |
Writing & Preference Kimi K3 leads
GLM-5.1: 66.9 (#31), Kimi K3: 76.6 (#4)
| Benchmark | GLM-5.1 | Kimi K3 |
|---|---|---|
| LMArena Text | 1461 | 1476 |
| LMArena Creative Writing | 1453 | 1454 |
| EQ-Bench Creative Writing | 1592 | 2082 |
| LMArena Multi-Turn | 1472 | 1488 |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is GLM-5.1 better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 47.8 on the Noometry Index. GLM-5.1 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-5.1 or Kimi K3?
GLM-5.1 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-5.1 or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 48.7 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 200K.
How many benchmarks do GLM-5.1 and Kimi K3 share?
36 benchmarks have published results for both models. GLM-5.1 has 41 scored results on Noometry and Kimi K3 has 53.