GLM-5.3-Flash vs Step-3.7-Flash
Specs, VRAM requirements and download trends — updated 24 October 2026.
stepfun-ai · B · Open weights
Params
201.4 B
Context
—
30d
17 K
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | GLM-5.3-Flash | Step-3.7-Flash |
|---|---|---|
| Deployment | Open weights · also via API | Open weights · also via API |
| Parameters | 321.3 B | 201.4 B |
| Architecture | Glm5NextForConditionalGeneration | Step3p7ForConditionalGeneration |
| Context length | 1,048,575 | 256,000 |
| Licence | mit | apache-2.0 |
| Commercial use | Allowed | Allowed |
| Languages | 2 | 1 |
| Downloads 30d | 6,319,699 | 16,549 |
| Downloads all time | 10.3 M | 406 K |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2026-08-25 | 2026-05-23 |
| API input price / 1M tokens | $0.15 | $0.20 |
| API output price / 1M tokens | $0.50 | $1.15 |
| Est. cost per 1M input + 1M output tokens | $0.65 | $1.35 |
| Cached input / 1M tokens | $0.03 | $0.04 |
| AA Intelligence Index | 41.8 | 19.5 |
| AA Coding Index | 71.5 | 39.6 |
| AA Agentic Index | 50.9 | — |
| Max output tokens | 943,717 | 230,400 |
| API providers | 34 | 2 |
| Tool calling | Yes | Yes |
| Reasoning mode | Yes | Yes |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
GLM-5.3-Flash
$0.65
Step-3.7-Flash
$1.35
GLM-5.3-Flash is 2.1× cheaper at list price.
Download trend
Daily snapshots, last 56 days (30 Aug – 24 Oct)
6.6 M17 K
GLM-5.3-Flash
Step-3.7-Flash
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
GLM-5.3-Flash · fp16
409.9 GB / 24 GB
Step-3.7-Flash · fp16
473.7 GB / 24 GB