DeepSeek-V4-Flash-0731 vs Kimi-K2-Thinking
Specs, VRAM requirements and download trends — updated 24 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 | Kimi-K2-Thinking |
|---|---|---|
| Deployment | Open weights · also via API | Open weights · also via API |
| Parameters | 304.2 B | 1,026.4 B |
| Architecture | DeepseekV4ForCausalLM | DeepseekV3ForCausalLM |
| Context length | 1,048,576 | 262,144 |
| Licence | mit | other |
| Commercial use | Allowed | Unknown |
| Languages | — | — |
| Downloads 30d | 4,348,313 | 47,623 |
| Downloads all time | 11.9 M | 2.0 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2026-07-31 | 2025-11-04 |
| API input price / 1M tokens | $0.044 | $0.60 |
| API output price / 1M tokens | $0.13 | $2.50 |
| Est. cost per 1M input + 1M output tokens | $0.18 | $3.10 |
| Cached input / 1M tokens | $0.0014 | $0.15 |
| AA Intelligence Index | 34.3 | 22 |
| AA Coding Index | 69.1 | — |
| AA Agentic Index | 41 | — |
| Max output tokens | 131,072 | 98,304 |
| API providers | 22 | 2 |
| Tool calling | Yes | Yes |
| Reasoning mode | Yes | Yes |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
DeepSeek-V4-Flash-0731
$0.18
Kimi-K2-Thinking
$3.10
DeepSeek-V4-Flash-0731 is 17.4× cheaper at list price.
Download trend
Daily snapshots, last 82 days (4 Aug – 24 Oct)
4.7 M49 K
DeepSeek-V4-Flash-0731
Kimi-K2-Thinking
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Kimi-K2-Thinking · fp16 @ 256K ctx
5,580.9 GB / 24 GB
DeepSeek-V4-Flash-0731 · fp16 @ 1024K ctx
6,024.3 GB / 24 GB