Step 3.7 FlashvsGLM-5V-Turbo
Across 6 shared benchmarks, Step 3.7 Flash leads overall: Step 3.7 Flash wins 4, GLM-5V-Turbo wins 0, with 2 ties and an average score difference of +2.23.
Step 3.7 Flash
StepFunAI · 2026-05-29 · Reasoning model
GLM-5V-Turbo
智谱AI · 2026-04-01 · Reasoning model
Step 3.7 Flash4 wins(67%)Ties2(0%)0 winsGLM-5V-Turbo
Benchmark scores
Grouped by capability, sorted by largest gap within each. 6 shared benchmarks.
Agent Level Benchmark
Step 3.7 Flash 1/2| Benchmark | Step 3.7 Flash | GLM-5V-Turbo | Diff |
|---|---|---|---|
| Terminal Bench Hard | 35.6070 / 244Thinking (With Tools) | 32.6093 / 244Thinking (With Tools) | +3 |
| τ²-Bench - Telecom | 98.504 / 264Thinking (With Tools) | 98.504 / 264Thinking (With Tools) | — |
General Evaluation
Even 1/1| Benchmark | Step 3.7 Flash | GLM-5V-Turbo | Diff |
|---|---|---|---|
| GPQA Diamond | 80.90225 / 462Thinking (No Tools) | 80.90225 / 462Thinking (No Tools) | — |
General Knowledge
Step 3.7 Flash 1/1| Benchmark | Step 3.7 Flash | GLM-5V-Turbo | Diff |
|---|---|---|---|
| CritPt | 2.30126 / 200Thinking (No Tools) | 0.60168 / 200Thinking (No Tools) | +1.70 |
Instruction Following
Step 3.7 Flash 1/1| Benchmark | Step 3.7 Flash | GLM-5V-Turbo | Diff |
|---|---|---|---|
| IF Bench | 67.3090 / 282Thinking (No Tools) | 61.10116 / 282Thinking (No Tools) | +6.20 |
Multimodal Understanding
Step 3.7 Flash 1/1| Benchmark | Step 3.7 Flash | GLM-5V-Turbo | Diff |
|---|---|---|---|
| MMMU-Pro | 75.3086 / 227Thinking (No Tools) | 72.80111 / 227Thinking (No Tools) | +2.50 |
Specs
| Field | Step 3.7 Flash | GLM-5V-Turbo |
|---|---|---|
| Publisher | StepFunAI | 智谱AI |
| Release date | 2026-05-29 | 2026-04-01 |
| Model type | Reasoning model | Reasoning model |
| Architecture | MoE | Dense |
| Parameters | 198B | Not available |
| Context length | 256K | 200K |
| Max output | Not available | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Step 3.7 Flash | GLM-5V-Turbo |
|---|---|---|
| Text input | ¥1.35 / 1M tokens | $1.2 / 1M tokens |
| Text output | ¥8.1 / 1M tokens | $4 / 1M tokens |
| Cache read | ¥0.27 / 1M tokens | $0.24 / 1M tokens |
Summary
- Step 3.7 Flashleads in:Agent Level Benchmark (1/2), General Knowledge (1/1), Instruction Following (1/1), Multimodal Understanding (1/1)
- Tied in:General Evaluation
On average across the 6 shared benchmarks, Step 3.7 Flash scores 2.23 higher.
Largest single-benchmark gap: IF Bench — Step 3.7 Flash 67.30 vs GLM-5V-Turbo 61.10 (+6.20).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.