Grouped by capability, sorted by largest gap within each. 14 shared benchmarks.
| Benchmark | GLM-5 | Kimi K2.5 | Diff |
|---|---|---|---|
| ARC-AGI | 44.7044 / 65Thinking (No Tools) | 65.3031 / 65Thinking (No Tools) | -20.60 |
| ARC-AGI-2 | 4.9043 / 58Thinking (No Tools) | 11.8035 / 58Thinking (No Tools) | -6.90 |
| GPQA Diamond | 8640 / 175Thinking (No Tools) | 87.6031 / 175Thinking (No Tools) | -1.60 |
| HLE | 50.4015 / 149thinking + 使用工具 | 50.2017 / 149Thinking (With Tools) | +0.20 |
| Benchmark | GLM-5 | Kimi K2.5 | Diff |
|---|---|---|---|
| FrontierMath - Tier 4 | 2.1056 / 80Normal (No Tools) | 4.2040 / 80Normal (No Tools) | -2.10 |
| IMO-AnswerBench | 82.5011 / 17Thinking (No Tools) | 81.8012 / 17Thinking (No Tools) | +0.70 |
| AIME 2026 | 92.707 / 14Thinking (No Tools) | 92.5010 / 14Thinking (No Tools) | +0.20 |
| Benchmark | GLM-5 | Kimi K2.5 | Diff |
|---|---|---|---|
| Claw Bench | 91.705 / 29Thinking (With Tools) | 81.7018 / 29Thinking (With Tools) | +10 |
| Pinch Bench | 86.4012 / 37Thinking (With Tools) | 84.8017 / 37Thinking (With Tools) | +1.60 |
| Benchmark | GLM-5 | Kimi K2.5 | Diff |
|---|---|---|---|
| BrowseComp | 75.9017 / 43thinking + 使用工具 | 60.6027 / 43Thinking (With Tools + Internet) | +15.30 |
| Benchmark | GLM-5 | Kimi K2.5 | Diff |
|---|---|---|---|
| Terminal Bench 2.0 | 61.1015 / 43thinking + 使用工具 | 50.8030 / 43Thinking (With Tools) | +10.30 |
| Benchmark | GLM-5 | Kimi K2.5 | Diff |
|---|---|---|---|
| SWE-bench Verified | 77.8018 / 103Thinking (No Tools) | 76.8022 / 103Thinking (With Tools) | +1 |
| Benchmark | GLM-5 | Kimi K2.5 | Diff |
|---|---|---|---|
| AA-LCR | 6312 / 13Thinking (No Tools) | 6510 / 13Thinking (No Tools) | -2 |
| Benchmark | GLM-5 | Kimi K2.5 | Diff |
|---|---|---|---|
| GDPval-AA | 4613 / 20Thinking (No Tools) | 4014 / 20Thinking (No Tools) | +6 |
| Field | GLM-5 | Kimi K2.5 |
|---|---|---|
| Publisher | 智谱AI | Moonshot AI |
| Release date | 2026-02-11 | 2026-01-27 |
| Model type | AI model | Multimodal model |
| Architecture | MoE | MoE |
| Parameters | 7440.0 | 10000.0 |
| Context length | 200K | 256K |
| Max output | 131072 | 16384 |
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GLM-5 | Kimi K2.5 |
|---|---|---|
| Text input | $1 / 1M tokens | 0.6 美元/100 万tokens |
| Text output | $3.2 / 1M tokens | 3 美元/100 万tokens |
| Cache read | Not public | 0.1 美元/100 万tokens |
| Cache write | $0.2 / 1M tokens | Not public |
On average across the 14 shared benchmarks, GLM-5 scores 0.86 higher.
Largest single-benchmark gap: ARC-AGI — GLM-5 44.70 vs Kimi K2.5 65.30 (-20.60).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.
