DeepSeek V4.1 Flash vs PrismML Ternary Bonsai 2 27B
Side by side from live data: prices, limits, who serves each model and how reliably, and how they do on our chess arena.
In short
- DeepSeek V4.1 Flash is 1.1× cheaper at a typical mix of 3 input tokens per output token.
- DeepSeek V4.1 Flash accepts longer prompts: 1M vs 262K tokens.
- DeepSeek V4.1 Flash output costs $0.18 per 1M via Decart, against the list price of $0.50.
| DeepSeek V4.1 Flash | Ternary Bonsai 2 27B | |
|---|---|---|
| Vendor | DeepSeek | PrismML |
| Released | 2026-09-10 | 2026-09-18 |
| Input / 1M | $0.045 | $0.075 |
| Output / 1M | $0.50 | $0.50 |
| Cached input / 1M | $0.0060 | $0.037 |
| Blended / 1M (3:1) | $0.159 | $0.181 |
| Context | 1M | 262K |
| Max output | 944K | 33K |
| Accepts | text, image | text, image |
| Reasoning | Yes | Yes |
| Providers on OpenRouter | 41 | 1 |
| Cheapest provider, output | $0.18 | $0.50 |
| Best provider uptime, 24 h | 100% | 100% |
| Chess arena rating | 1472 ±77 | not playing |
| Vendor API status | Operational | — |
Cost calculator
DeepSeek V4.1 Flash
$10.20 per month$0.34 per day
Ternary Bonsai 2 27B
$12.00 per month$0.40 per day
DeepSeek V4.1 Flash saves $1.80 a month on this workload.
At list prices per 1M tokens; prompt caching and batch discounts are not included.
Compare DeepSeek V4.1 Flash with
Compare Ternary Bonsai 2 27B with