DeepSeek-V4-Flash-0731 vs Hy3
Specs, VRAM requirements and download trends — updated 23 October 2026.
deepseek-ai · A · Open weights
Params
304.2 B
Context
1,048,576
30d
4.3 M
mit · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | DeepSeek-V4-Flash-0731 | Hy3 |
|---|---|---|
| Deployment | Open weights · also via API | Open weights · also via API |
| Parameters | 304.2 B | 298.8 B |
| Architecture | DeepseekV4ForCausalLM | HYV3ForCausalLM |
| Context length | 1,048,576 | 262,144 |
| Licence | mit | apache-2.0 |
| Commercial use | Allowed | Allowed |
| Languages | — | — |
| Downloads 30d | 4,297,931 | 797,888 |
| Downloads all time | 11.7 M | 842 K |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2026-07-31 | 2026-07-02 |
| API input price / 1M tokens | $0.44 | $0.13 |
| API output price / 1M tokens | $1.32 | $0.53 |
| Est. cost per 1M input + 1M output tokens | $1.76 | $0.66 |
| Cached input / 1M tokens | $0.014 | $0.033 |
| AA Intelligence Index | 34.3 | — |
| AA Coding Index | 69.1 | — |
| AA Agentic Index | 41 | — |
| Max output tokens | 131,072 | 128,000 |
| API providers | 23 | 5 |
| Tool calling | Yes | Yes |
| Reasoning mode | Yes | Yes |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
DeepSeek-V4-Flash-0731
$1.76
Hy3
$0.66
Hy3 is 2.7× cheaper at list price.
Download trend
Daily snapshots, last 81 days (4 Aug – 23 Oct)
4.7 M188 K
DeepSeek-V4-Flash-0731
Hy3
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Hy3 · fp16 @ 256K ctx
2,092.0 GB / 24 GB
DeepSeek-V4-Flash-0731 · fp16 @ 1024K ctx
6,024.3 GB / 24 GB