Data points: 49
Model compare
The readout for Devstral 2 and GLM-5, before the detailed comparison sheet.
Weighted outcome: GLM-5. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.
Decision read
GLM-5
GLM-5 has the higher weighted result; Model A / B score 10 to 90.
Evidence depth
49 data points
Includes 7 benchmark rows, 0 audit samples, and 5 provider examples.
Selection signal
Start with GLM-5
The charts below split 12 high-signal samples across speed, scores, and audit health.
Switch either side of this report to compare another model with the same LMSpeed data pipeline.
Select a different model to open a new comparison URL.
This report only uses LMSpeed data for Devstral 2 and GLM-5: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.
| Model compare | Devstral 2 | GLM-5 |
|---|---|---|
| Overall leader | Contender | Leading |
| Weighted overall score | 10.0 pts | 90.0 pts |
| Benchmark category leads | 0 categories | 3 categories |
| Operational advantages | Average speed, First-token latency | Free providers, Provider coverage |
| Context window | No data | 202.8K tokens |
| Max output | No data | 202.8K tokens |
| Modalities | No data | Input Text Output Text |
| Features |
The overall result weights benchmark capability categories at 80% and price, API speed/latency, and availability at 20%. Recent test volume does not affect the winner, and missing benchmark categories are excluded.
| Model compare | Devstral 2 | GLM-5 |
|---|---|---|
| Developer | No data | Z.ai |
| Released | No data | Feb 2026 |
| Parameters | No data | No data |
| Tokenizer | No data | Other |
| Knowledge cutoff | No data | No data |
| OpenRouter ID | No data | z-ai/glm-5 |
| References | No data | No data |
This report only uses LMSpeed data for Devstral 2 and GLM-5: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.
Devstral 2
Devstral 2 has these operational advantages: Average speed, First-token latency.
GLM-5
GLM-5 is stronger in benchmark categories (Coding, Reasoning, Math) and operational dimensions (Free providers, Provider coverage).
Third-party benchmark profile synced into LMSpeed; only metrics available for both models are shown.
Compare benchmark category scores on a 0-100 scale. Select a category to inspect the gap.
Avg. score
Devstral 2
45.2
Avg. score
GLM-5
48.5
GLM-5
GLM-5 leads by 6.5
GLM-5 leads by 4.0
GLM-5
GLM-5 leads by 8.3
GLM-5
GLM-5
GLM-5
Metric-level scores with benchmark source, rank depth, confidence, error, and evaluation date where available.
Devstral 2
winner0.71 s
Rank #25/139 · confidence 4
GLM-5
63.52 s
Rank #126/139 · confidence 4
Devstral 2
winner0.71 s
Rank #34/139 · confidence 4
GLM-5
0.76 s
Rank #38/139 · confidence 4
Devstral 2
winner60.4 tok/s
Rank #95/139 · confidence 4
GLM-5
49.5 tok/s
Rank #113/139 · confidence 4
Devstral 2
76.2%
Rank #80/125 · confidence 4
GLM-5
winner85.7%
Rank #23/125 · confidence 4 · eval date 2026-03-01
Devstral 2
3.6%
Rank #172/186 · confidence 4
GLM-5
winner27.2%
Rank #37/186 · confidence 4
Devstral 2
59.4%
Rank #141/187 · confidence 4
GLM-5
winner82.0%
Rank #66/187 · confidence 4
Devstral 2
33.1%
Rank #120/184 · confidence 4
GLM-5
winner46.2%
Rank #31/184 · confidence 4
Latest completed audits from shared providers, with four safety and integrity score groups plus report links.
| Provider | Devstral 2 | GLM-5 |
|---|---|---|
| No completed audits are available from shared providers yet. | ||
Speed aggregates and input/output pricing share each provider row for real API selection and migration cost checks.
| Provider | Devstral 2 | GLM-5 |
|---|---|---|
10 tests | Devstral 2 speed / latency 71 tok/s / 1341ms input / output No data | GLM-5 speed / latency 102 tok/s / 8791ms input / output No data |
10 tests | Devstral 2 speed / latency N/A / N/A input / output No data | GLM-5 speed / latency 64 tok/s / 21679ms input / output No data |
10 tests | Devstral 2 speed / latency N/A / N/A input / output No data | GLM-5 speed / latency 41 tok/s / 18111ms input / output No data |
0 tests | Devstral 2 speed / latency N/A / N/A input / output No data | GLM-5 speed / latency N/A / N/A input / output No data |
0 tests | Devstral 2 speed / latency N/A / N/A input / output No data | GLM-5 speed / latency N/A / N/A input / output No data |
Weighted outcome: GLM-5. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.
| None listed |
Text inputText outputTool callingStructured outputsJSON modeReasoning |
