DeepSeek-V4-Flash vs Qwen3-235B-A22B-Instruct-2507

Specs, VRAM requirements and download trends — updated 22 September 2026.

deepseek-ai · A
Params
158.1 B
Context
1,048,576
30d
1.5 M
mit · commercial OK
Qwen · B
Params
235.1 B
Context
262,144
30d
197 K
apache-2.0 · commercial OK

Specification comparison

Differences are highlighted; identical values are muted.

Specification DeepSeek-V4-Flash Qwen3-235B-A22B-Instruct-2507
Parameters 158.1 B 235.1 B
Architecture DeepseekV4ForCausalLM Qwen3MoeForCausalLM
Context length 1,048,576 262,144
Licence mit apache-2.0
Commercial use Allowed Allowed
Languages
Downloads 30d 1,524,356 197,322
Downloads all time 12.0 M 1.8 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated No No
First seen on the Hub 2026-04-22 2025-07-21

Download trend

57 days of snapshots recorded — chart in 0 days.

12.0 M
DeepSeek-V4-Flash · all-time downloads
1.8 M
Qwen3-235B-A22B-Instruct-2507 · all-time downloads

VRAM side by side

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

Qwen3-235B-A22B-Instruct-2507 · fp16 @ 256K ctx 1,646.2 GB / 24 GB
❌ Won’t fit
DeepSeek-V4-Flash · fp16 @ 1024K ctx 3,211.0 GB / 24 GB
❌ Won’t fit

Adjacent comparisons