GLM 5.1vsGLM-5
Across 14 shared benchmarks, GLM 5.1 leads overall: GLM 5.1 wins 10, GLM-5 wins 4, with 0 ties and an average score difference of +3.40.
GLM 5.110 wins(71%)(29%)4 winsGLM-5
Benchmark scores
Grouped by capability, sorted by largest gap within each. 14 shared benchmarks.
Agent Level Benchmark
GLM-5 2/3| Benchmark | GLM 5.1 | GLM-5 | Diff |
|---|---|---|---|
| τ³-Banking | 13.60115 / 164Thinking (With Tools) | 9.79130 / 164Thinking (With Tools) | +3.81 |
| Terminal Bench Hard | 35.6070 / 244Normal (With Tools) | 39.4053 / 244Normal (With Tools) | -3.80 |
| τ²-Bench - Telecom | 97.1017 / 264Normal (With Tools) | 97.4016 / 264Normal (With Tools) | -0.30 |
General Knowledge
GLM 5.1 3/3| Benchmark | GLM 5.1 | GLM-5 | Diff |
|---|---|---|---|
| HLE | 27.90226 / 563Normal (No Tools) · Text only | 7.60427 / 563Normal (No Tools) · Text only | +20.30 |
| CritPt | 4.60102 / 200Thinking (No Tools) | 2127 / 200Thinking (No Tools) | +2.60 |
| LiveBench | 70.1837 / 117Normal (No Tools) | 68.8543 / 117Normal (No Tools) | +1.33 |
Math and Reasoning
GLM 5.1 2/2| Benchmark | GLM 5.1 | GLM-5 | Diff |
|---|---|---|---|
| AIME 2026 | 95.3012 / 29Thinking (No Tools) | 92.7018 / 29Thinking (No Tools) | +2.60 |
| IMO-AnswerBench | 83.8014 / 24Thinking (No Tools) | 82.5017 / 24Thinking (No Tools) | +1.30 |
AI Agent - Tool Usage
GLM 5.1 1/1| Benchmark | GLM 5.1 | GLM-5 | Diff |
|---|---|---|---|
| Terminal Bench 2.0 | 63.5013 / 48Thinking (With Tools) | 61.1018 / 48Thinking (With Tools) | +2.40 |
Commonsense Reasoning
GLM 5.1 1/1| Benchmark | GLM 5.1 | GLM-5 | Diff |
|---|---|---|---|
| SimpleBench | 55.1042 / 93Normal (No Tools) | 53.2045 / 93Normal (No Tools) | +1.90 |
General Evaluation
GLM 5.1 1/1| Benchmark | GLM 5.1 | GLM-5 | Diff |
|---|---|---|---|
| GPQA Diamond | 83.90180 / 462Normal (No Tools) | 66.60349 / 462Normal (No Tools) | +17.30 |
Instruction Following
GLM-5 1/1| Benchmark | GLM 5.1 | GLM-5 | Diff |
|---|---|---|---|
| IF Bench | 52148 / 282Normal (No Tools) | 55.20132 / 282Normal (No Tools) | -3.20 |
Long Context
GLM 5.1 1/1| Benchmark | GLM 5.1 | GLM-5 | Diff |
|---|---|---|---|
| AA-LCR | 53.30134 / 170Normal (No Tools) | 43.70151 / 170Normal (No Tools) | +9.60 |
Writing and Creative Capabilities
GLM-5 1/1| Benchmark | GLM 5.1 | GLM-5 | Diff |
|---|---|---|---|
| Creative Writing | 1,58940 / 106Normal (No Tools) | 1,59839 / 106Normal (No Tools) | -8.30 |
Specs
| Field | GLM 5.1 | GLM-5 |
|---|---|---|
| Publisher | 智谱AI | 智谱AI |
| Release date | 2026-03-27 | 2026-02-11 |
| Model type | Reasoning model | Chat model |
| Architecture | MoE | MoE |
| Parameters | 754B | 744B |
| Context length | 200K | 200K |
| Max output | 125K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GLM 5.1 | GLM-5 |
|---|---|---|
| Text input | $1.4 / 1M tokens | $1 / 1M tokens |
| Text output | $4.4 / 1M tokens | $3.2 / 1M tokens |
| Cache read | $4.4 / 1M tokens | Not public |
| Cache write | $0.26 / 1M tokens | $0.2 / 1M tokens |
Summary
- GLM 5.1leads in:General Knowledge (3/3), Math and Reasoning (2/2), AI Agent - Tool Usage (1/1), Commonsense Reasoning (1/1), General Evaluation (1/1), Long Context (1/1)
- GLM-5leads in:Agent Level Benchmark (2/3), Instruction Following (1/1), Writing and Creative Capabilities (1/1)
On average across the 14 shared benchmarks, GLM 5.1 scores 3.40 higher.
Largest single-benchmark gap: HLE — GLM 5.1 27.90 vs GLM-5 7.60 (+20.30).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.