DeepSeek-V3 vs DeepSeek-V3.2-AWQ
Specs, VRAM requirements and download trends — updated 14 August 2026.
QuantTrio · B · Open weights
Params
685.4 B
Context
163,840
30d
168 K
mit · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | DeepSeek-V3 | DeepSeek-V3.2-AWQ |
|---|---|---|
| Deployment | Open weights · also via API | Open weights |
| Parameters | 684.5 B | 685.4 B |
| Architecture | DeepseekV3ForCausalLM | DeepseekV32ForCausalLM |
| Context length | 163,840 | 163,840 |
| Licence | — | mit |
| Commercial use | Unknown | Allowed |
| Languages | — | — |
| Downloads 30d | 1,487,156 | 167,983 |
| Downloads all time | 23.5 M | 1.5 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2024-12-25 | 2025-12-03 |
| API input price / 1M tokens | $0.32 | not on OpenRouter |
| API output price / 1M tokens | $0.89 | not on OpenRouter |
| Est. cost per 1M input + 1M output tokens | $1.21 | not on OpenRouter |
| Cached input / 1M tokens | — | not on OpenRouter |
| AA Intelligence Index | — | not on OpenRouter |
| AA Coding Index | — | not on OpenRouter |
| AA Agentic Index | — | not on OpenRouter |
| Max output tokens | 16,384 | not on OpenRouter |
| API providers | 2 | not on OpenRouter |
| Tool calling | Yes | not on OpenRouter |
| Reasoning mode | No | not on OpenRouter |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
DeepSeek-V3
$1.21
Download trend
Daily snapshots, last 90 days (13 Aug – 10 Nov)
1.7 M168 K
DeepSeek-V3
DeepSeek-V3.2-AWQ
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
DeepSeek-V3.2-AWQ · fp16 @ 160K ctx
2,454.7 GB / 24 GB
DeepSeek-V3 · fp16 @ 160K ctx
2,811.5 GB / 24 GB