DeepSeek-V3.2 vs Qwen3.8-Flash-Next-NVFP4-GPTQ-Strata
Specs, VRAM requirements and download trends — updated 25 October 2026.
deepseek-ai · A · Open weights
Params
685.4 B
Context
163,840
30d
5.6 M
mit · commercial OK
Maximilian228 · B · Open weights
Params
180.0 B
Context
—
30d
245 K
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | DeepSeek-V3.2 | Qwen3.8-Flash-Next-NVFP4-GPTQ-Strata |
|---|---|---|
| Deployment | Open weights · also via API | Open weights |
| Parameters | 685.4 B | 180.0 B |
| Architecture | DeepseekV32ForCausalLM | — |
| Context length | 163,840 | — |
| Licence | mit | other |
| Commercial use | Allowed | Unknown |
| Languages | — | 2 |
| Downloads 30d | 5,571,243 | 244,900 |
| Downloads all time | 30.7 M | 245 K |
| Quantizations on the Hub | SAFETENSORS | GGUF |
| Gated | No | No |
| First seen on the Hub | 2025-12-01 | 2026-10-09 |
| API input price / 1M tokens | $0.28 | not on OpenRouter |
| API output price / 1M tokens | $0.42 | not on OpenRouter |
| Est. cost per 1M input + 1M output tokens | $0.70 | not on OpenRouter |
| Cached input / 1M tokens | $0.028 | not on OpenRouter |
| AA Intelligence Index | 21.5 | not on OpenRouter |
| AA Coding Index | 44.2 | not on OpenRouter |
| AA Agentic Index | — | not on OpenRouter |
| Max output tokens | 65,536 | not on OpenRouter |
| API providers | 11 | not on OpenRouter |
| Tool calling | Yes | not on OpenRouter |
| Reasoning mode | Yes | not on OpenRouter |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
DeepSeek-V3.2
$0.70
Download trend
Daily snapshots, last 90 days (28 Jul – 25 Oct)
5.6 M146 K
DeepSeek-V3.2
Qwen3.8-Flash-Next-NVFP4-GPTQ-Strata
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
DeepSeek-V3.2 · fp16 @ 160K ctx
2,815.1 GB / 24 GB