Qwen3.8-27B-iMatrix-NVFP4-MTP-GGUF vs Step-3.7-Flash

Specs, VRAM requirements and download trends — updated 24 October 2026.

cdiamond · A · Open weights

Params

—

Context

—

30d

2.4 M

apache-2.0 · commercial OK

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	Qwen3.8-27B-iMatrix-NVFP4-MTP-GGUF	Step-3.7-Flash
Deployment	Open weights	Open weights · also via API
Parameters	—	201.4 B
Architecture	—	Step3p7ForConditionalGeneration
Context length	—	256,000
Licence	apache-2.0	apache-2.0
Commercial use	Allowed	Allowed
Languages	—	1
Downloads 30d	2,369,313	16,549
Downloads all time	7.5 M	406 K
Quantizations on the Hub	GGUF	SAFETENSORS
Gated	No	No
First seen on the Hub	2026-08-17	2026-05-23
API input price / 1M tokens	not on OpenRouter	$0.20
API output price / 1M tokens	not on OpenRouter	$1.15
Est. cost per 1M input + 1M output tokens	not on OpenRouter	$1.35
Cached input / 1M tokens	not on OpenRouter	$0.04
AA Intelligence Index	not on OpenRouter	19.5
AA Coding Index	not on OpenRouter	39.6
AA Agentic Index	not on OpenRouter	—
Max output tokens	not on OpenRouter	230,400
API providers	not on OpenRouter	2
Tool calling	not on OpenRouter	Yes
Reasoning mode	not on OpenRouter	Yes

Estimated cost of 1M input + 1M output tokens via OpenRouter

Step-3.7-Flash $1.35

Daily snapshots, last 53 days (2 Sep – 24 Oct)

5.0 M17 K

Qwen3.8-27B-iMatrix-NVFP4-MTP-GGUF Step-3.7-Flash

On RTX 4090 · 24 GB · 8K context unless noted

Step-3.7-Flash · fp16 473.7 GB / 24 GB

❌ Won’t fit