Step 3.7 Flash leads on the shared-benchmark average
Ahead on 4 of 6 shared benchmarks, averaging 2.2 points higher
Summarised only from the 6 percentage-scale benchmarks scored by every selected model; details are below.
“Best available” takes each model’s best recorded non-parallel result in the metric’s stated direction per benchmark, so it may combine modes into a virtual configuration that does not exist. Read it with the mode breakdown.
Step 3.7 Flash
StepFunAI
Benchmark-by-benchmark comparison. Changing the thinking mode or tool filters updates the chart and table below.
“Best available” picks the best non-parallel result in the metric’s stated direction separately for each benchmark. The resulting series can combine several reasoning levels and is not one reproducible runtime configuration. Choose a mode filter to compare like-for-like runs.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Each axis is the mean percentage score of one benchmark domain. It is an average, not a capability rating.
Relative edge: 综合评估 +9.4 / Relative gap: none clear
Relative edge: none clear / Relative gap: 综合评估 -9.4
Method: for each model and benchmark, all scores in the current mode scope are averaged (not the best score), then those benchmark scores are averaged within each domain. Only benchmarks scored on a 0-100 scale by at least two of the selected models count — Elo and rating-scale benchmarks such as Codeforces or Arena are excluded, because averaging a 1500 rating with an 85% accuracy produces a meaningless number. Missing values are not counted as zero, and the averages are unweighted, so domains with harder benchmarks read lower.
Every model and runtime mode, benchmark by benchmark. Values are comparable along a row, not between different benchmarks.
7 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | Step 3.7 Flash | GLM-5V-Turbo |
|---|---|---|
2.30Thinking Enabled | 0.60Thinking Enabled | |
47.20Thinking Enabled ｜ Tools | 17.10Thinking Enabled | |
80.90Thinking Enabled | 80.90Thinking Enabled | |
75.30Thinking Enabled | 72.80Thinking Enabled | |
35.60Thinking Enabled ｜ Tools | 32.60Thinking Enabled ｜ Tools | |
98.50Thinking Enabled ｜ Tools | 98.50Thinking Enabled ｜ Tools | |
67.30Thinking Enabled | 61.10Thinking Enabled |
Official list prices per model API, split by input and output. Unit: USD per 1M tokens.
Architecture, licensing and API modalities. "Not provided" means the field is missing from our database.
| Features & specs | Step 3.7 FlashStepFunAI | GLM-5V-Turbo智谱AI |
|---|---|---|
Core specsRelease | 2026-05-29 | 2026-04-01 |
Context length | 256K | 200K |
Total parameters | 198B | — |
Active parameters | 11B | N/A |
Max output length | Not provided | 131,072 tokens |
Architecture | MoE (mixture of experts) | Undisclosed |
Runtime modes | 关闭深度 | 关闭开启 |
Availability & licensingCode availability | Available · Apache 2.0 | Not public |
Weight availability | Available · Apache 2.0 | Not public |
Use & commercial terms | 免费商用授权 | Official service only; subject to provider terms |
Local deploymentWeight size | 403 GB（BF16 Safetensors） | 未知 |
VRAM for weights | ≈ 403 GB (weights only, excludes KV cache) | Not provided |
Weights | Hugging Face | Not provided |
Source repo | GitHub | Not provided |
API modality supportText Input/Output | Input：YesOutput：Yes | Input：YesOutput：Yes |
Image Input/Output | Input：YesOutput：No | Input：YesOutput：No |
Video Input/Output | Input：YesOutput：No | Input：YesOutput：No |
ResourcesPaper / report | Step 3.7 Flash — A high-efficiency Flash model for Real-World Agents | GLM-5V-Turbo Model Guide |
GLM-5V-Turbo
智谱AI
