DeepSeek-V3 vs Hermes-4-405B
Specs, VRAM requirements and download trends — updated 24 October 2026.
NousResearch · B · Open weights
Params
405.9 B
Context
131,072
30d
1 K
llama3 licence · conditional
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | DeepSeek-V3 | Hermes-4-405B |
|---|---|---|
| Deployment | Open weights · also via API | Open weights · also via API |
| Parameters | 684.5 B | 405.9 B |
| Architecture | DeepseekV3ForCausalLM | LlamaForCausalLM |
| Context length | 163,840 | 131,072 |
| Licence | — | llama3 |
| Commercial use | Unknown | Conditional |
| Languages | — | 1 |
| Downloads 30d | 1,344,492 | 1,216 |
| Downloads all time | 22.5 M | 8 K |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2024-12-25 | 2025-08-06 |
| API input price / 1M tokens | $0.26 | $1.00 |
| API output price / 1M tokens | $1.03 | $3.00 |
| Est. cost per 1M input + 1M output tokens | $1.29 | $4.00 |
| Cached input / 1M tokens | — | — |
| AA Intelligence Index | — | 7.5 |
| AA Coding Index | — | — |
| AA Agentic Index | — | — |
| Max output tokens | 16,000 | 117,964 |
| API providers | 2 | 1 |
| Tool calling | Yes | No |
| Reasoning mode | No | Yes |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
DeepSeek-V3
$1.29
Hermes-4-405B
$4.00
DeepSeek-V3 is 3.1× cheaper at list price.
Download trend
Daily snapshots, last 89 days (28 Jul – 24 Oct)
1.5 M1 K
DeepSeek-V3
Hermes-4-405B
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Hermes-4-405B · fp16 @ 128K ctx
1,867.4 GB / 24 GB
DeepSeek-V3 · fp16 @ 160K ctx
2,811.5 GB / 24 GB