Model comparison
GLM-5 vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 46.1 on the Noometry Index. GLM-5 costs 3.9× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. GLM-5 scores higher in 0 categories and Kimi K3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 27.6.
- The biggest single-benchmark swing is ARC-AGI-2: 4.9% for GLM-5 and 60.4% for Kimi K3.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 205K.
Side by side
| GLM-5 | Kimi K3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Moonshot AI |
| Noometry Index | 46.1 | 59.5 |
| Released | 2026-02-11 | 2026-07-16 |
| Weights | Open | Open |
| Context window | 205K | 1.05M |
| Max output | 131K | 1.05M |
| Input $ / M tokens | $1 | $3 |
| Output $ / M tokens | $3.20 | $15 |
| Results tracked | 45 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
GLM-5: 49.0 (#52), Kimi K3: 61.0 (#10)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| LMArena WebDev | 1434 | 1654 |
| WeirdML | 48.2% | 82.6% |
| LMArena Coding | 1461 | 1508 |
| ALE-Bench | 765.62 | 1,524 |
| SWE-bench Verified | 72.1% | — |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| SWE-bench Verified (bash only) | 72.8% | — |
| SWE-bench Multilingual | 69.7% | — |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
Agentic & Tool Use Kimi K3 leads
GLM-5: 31.1 (#71), Kimi K3: 41.8 (#20)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| τ²-bench Banking | 9.8% | 37.1% |
| Vending-Bench 2 | 4,432 | 5,165 |
| Terminal-Bench | 52.4% | — |
| APEX-Agents | — | 50.6% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
Reasoning Kimi K3 leads
GLM-5: 27.6 (#116), Kimi K3: 63.0 (#17)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 4.9% | 60.4% |
| SimpleBench | 53.2% | 60.7% |
| NYT Connections (extended) | 74.8% | 93.6% |
| ARC-AGI-1 | 44.7% | 94.5% |
| Chess Puzzles | 10% | 39% |
| LMArena Hard Prompts | 1452 | 1496 |
| Epoch Capabilities Index | 145.83 | 157.45 |
| ForecastBench | 61 | 61.1 |
| Kagi LLM Benchmark | 75% | — |
| CritPt | — | 23.4% |
| Mystery Game Puzzles | — | 26% |
| DTBench | — | 91.2% |
| LMCA | — | 52.7% |
| Surface Evolver Bench | — | 95% |
Math Kimi K3 leads
GLM-5: 46.4 (#71), Kimi K3: 74.2 (#16)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| MathArena Final-Answer Competitions | 65.7% | 87.8% |
| OTIS Mock AIME 2024-2025 | 80% | 97.2% |
| LMArena Math | 1440 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| ProofBench | — | 87% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Kimi K3 leads
GLM-5: 52.3 (#64), Kimi K3: 63.2 (#21)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| GPQA Diamond | 87.8% | 93.1% |
| LMArena Expert | 1454 | 1521 |
| SimpleQA Verified | — | 50.6% |
| Vectara Hallucination Rate | 10.1% | — |
Multimodal Not comparable
GLM-5: —, Kimi K3: 37.8 (#70)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Kimi K3 leads
GLM-5: 53.7 (#58), Kimi K3: 56.3 (#21)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1430 | 1466 |
| LMArena Chinese | 1511 | 1529 |
| LMArena French | 1455 | 1491 |
| LMArena German | 1445 | 1488 |
| LMArena Japanese | 1416 | 1487 |
| LMArena Korean | 1423 | 1458 |
| LMArena Russian | 1436 | 1482 |
| LMArena Spanish | 1454 | 1472 |
Instruction Following Kimi K3 leads
GLM-5: 75.2 (#67), Kimi K3: 77.7 (#14)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1428 | 1483 |
Long Context Kimi K3 leads
GLM-5: 44.7 (#60), Kimi K3: 45.8 (#29)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1446 | 1494 |
| CL-bench | 18.7% | — |
Writing & Preference Kimi K3 leads
GLM-5: 66.0 (#38), Kimi K3: 76.6 (#4)
| Benchmark | GLM-5 | Kimi K3 |
|---|---|---|
| LMArena Text | 1446 | 1476 |
| LMArena Creative Writing | 1439 | 1454 |
| EQ-Bench Creative Writing | 1601 | 2082 |
| LMArena Multi-Turn | 1456 | 1488 |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is GLM-5 better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 46.1 on the Noometry Index. GLM-5 costs 3.9× 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 or Kimi K3?
GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; Kimi K3 lists at $3 and $15.
Is GLM-5 or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 49.0 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 205K.
How many benchmarks do GLM-5 and Kimi K3 share?
33 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Kimi K3 has 53.