GLM-5 vs Kimi K3
GLM-5 
Kimi K3 Verdict
- Kimi K3 leads on quality: 66.5 vs 59.1.
- GLM-5 is stronger in instruction following, reasoning.
- Kimi K3 is stronger in agents & tools, coding, composite, human preference, knowledge, long context, maths, multimodal.
- GLM-5 is 3.9× cheaper ($1.55 vs $6.00 per 1M blended).
- GLM-5 streams 1.7× faster (66 vs 38 tokens per second).
| Metric | GLM-5 | Kimi K3 |
|---|---|---|
| BenchLeader Index | 59.1 | 66.5 |
| Agents & tools score | 56.8 | 67.4 |
| Coding score | 52.9 | 65.0 |
| Composite score | 64.3 | 74.2 |
| Human preference score | 65.6 | 71.2 |
| Instruction following score | 70.5 | – |
| Knowledge score | 64.3 | 68.2 |
| Long context score | 64.3 | 71.1 |
| Maths score | 53.1 | 78.1 |
| Reasoning score | 51.5 | – |
| Multimodal score | – | 65.0 |
| Blended price $/M | $1.55 | $6.00 |
| Output speed | 66 tok/s | 38 tok/s |
| Time to first answer | 48.7 s | 56.3 s |
| Context window | 205k | 1.0M |
| GPQA Diamond | 87.8% | – |
| OTIS Mock AIME | 80.0% | – |
| SWE-bench Verified (Epoch) | 72.1% | – |
| Terminal-Bench | 52.4% | – |
| SimpleBench | 53.2% | – |
| SciCode | – | 58.7% |
| WeirdML | 48.2% | – |
| APEX-Agents | 17.2% | 39.3% |
| FrontierCode | – | 44.2% |
| ProofBench | – | 87.0% |
| Epoch Capabilities Index | 145.9 | 157.6 |
| LMArena Text | 1458 | – |
| LMArena Hard Prompts | 1478 | – |
| LMArena Coding | 1498 | – |
| LMArena WebDev | 1436 | – |
| LiveBench | – | 79.2% |
| LiveBench Reasoning | – | 90.7% |
| LiveBench Coding | – | 81.5% |
| LiveBench Agentic Coding | – | 62.2% |
| LiveBench Mathematics | – | 84.4% |
| LiveBench Data Analysis | – | 78.7% |
| LiveBench Language | – | 85.5% |
| AA Intelligence Index | 27.9 | 43.8 |
| IFBench | 72.3% | – |
| AA-LCR | 75.7% | 88.7% |
| MMMU-Pro | – | 80.5% |
| AA-Omniscience | 0.3 | 19.7 |
| Terminal-Bench Hard | 43.2% | – |
| GPQA Diamond (AA) | 82.0% | 93.5% |
| Humanity's Last Exam (AA) | 29.3% | 46.9% |
| SciCode (AA) | – | 59.5% |
| τ²-Bench Telecom (AA) | 98.3% | – |
| LegalBench | – | 86.0% |
| CorpFin | – | 71.6% |
| TaxEval | – | 75.7% |
| Terminal-Bench 2.1 (Vals) | – | 80.9% |
| SWE-bench (Vals) | – | 93.4% |
| GPQA Diamond (Vals) | – | 92.9% |
| Vals Index | – | 57.8 |
| AIME 2026 | 96.7% | – |
| HMMT February 2026 | 86.4% | – |
| MathArena Apex | 10.9% | – |
| EQ-Bench 4 | – | 1339 |
| Kagi LLM Benchmark | 51.7% | – |
| ARC-AGI-1 | 44.7% | – |
| ARC-AGI-2 | 4.9% | – |
Data as of 2026-09-10. Best configuration of each model; every score links to its source on the model pages.
GLM-5 vs Kimi K3: questions
- Is GLM-5 better than Kimi K3?
- Kimi K3 leads on quality: 66.5 vs 59.1. The BenchLeader Index combines every independent quality benchmark; Kimi K3 is ahead overall as of 2026-09-10, but check the category scores for your use.
- Is GLM-5 better than Kimi K3 for coding?
- Kimi K3 scores higher in coding (65 vs 53 on the category index, where 50 is average).
- Is GLM-5 better than Kimi K3 for agentic tasks?
- Kimi K3 scores higher in agentic tasks (67 vs 57 on the category index, where 50 is average).
- Which is cheaper, GLM-5 or Kimi K3?
- GLM-5 is cheaper: $1.55 against $6.00 per million tokens, blended at three input tokens per output token.
- Which is faster, GLM-5 or Kimi K3?
- GLM-5 streams faster: 66 against 38 output tokens per second.
- Which has the larger context window?
- Kimi K3 accepts more context: 1.0M against 205k tokens.