Models / Compare /DeepSeek V4 Flash vs Kimi K2.6
Head to head
DeepSeek V4 Flash vs Kimi K2.6
Compared on live catalog data — pricing synced daily, benchmark scores with provenance.
The short version
- DeepSeek V4 Flash is cheaper to run hosted: $0.28/M output vs $2.48/M for Kimi K2.6 (8.9× difference at the cheapest tracked provider).
- DeepSeek V4 Flash takes 1M tokens of context vs 262K for Kimi K2.6.
- DeepSeek V4 Flash has downloadable weights (OPEN) so you can self-host it; we hold no downloadable copy of Kimi K2.6, so it runs through a provider's API.
- Kimi K2.6 scores 70.54 via Thinking on LiveBench vs 65.48 for DeepSeek V4 Flash.
- Kimi K2.6 scores 46.92 via Thinking on LiveBench Agentic Coding vs 37.63 for DeepSeek V4 Flash.
| Spec | ||
|---|---|---|
| Vendor | DeepSeek | Moonshot AI |
| Parameters | 291B / 37B | 1.1T / 32B |
| 1M | 262K | |
| Weights | OPEN | OPEN* |
| License | MIT License | Custom licence |
| Modality | text->text | text+image->text |
| Released | Apr 22, 2026 | Apr 14, 2026 |
| 183.4 GBest | 667.5 GBest | |
| Cheapest hosted | $0.14 / $0.28 per 1M (Novita AI) | $0.59 / $2.48 per 1M (OpenRouter) |
| Tracked providers | 32 | 26 |
DeepSeek V4 Flash — our take
DeepSeek V4 Flash is a large downloadable text model built for long documents, with a permissive MIT licence and a one-million-token request limit. Its measured web-development coding score is the strongest of its seven tracked skills, while creative writing is the weakest.
full breakdown →Kimi K2.6 — our take
Kimi K2.6 is a large mixture-of-experts model from Moonshot AI that accepts text and images and handles up to 262,144 tokens in a single request. It is the strongest coder in our measured set, with a coding score well above its general chat rating, though its creative writing lags behind both.
full breakdown →Need more columns? Use theinteractive comparison to add up to four models.