PrismML: Ternary Bonsai 2 27B API Pricing
PrismML: Ternary Bonsai 2 27B costs $0.075 per 1M input tokens and $0.500 per 1M output tokens (cache reads: $0.037). Context window: 262K. Prices verified 2026-10-21 · source
Rebuilt daily from our snapshot archive — the chart appears once a model has 5+ snapshots of history.
High adoption means battle-tested weights, more fine-tunes/quants, and faster bug fixes. View on HuggingFace →
From our own daily snapshot archive — independent of provider announcements.
Token economics
| Token type | Price | vs base input |
|---|---|---|
| Standard input | $0.075 | — |
| Cache read | $0.037 | −50% |
| Output | $0.500 | — |
Prompt caching stores repeated context server-side: reads cost far less, but writes carry a one-time premium.
Higher effort = deeper thinking = more reasoning tokens billed as output.
What would PrismML: Ternary Bonsai 2 27B cost you per month?
| Workload | Requests/mo | Avg tokens (in/out) | Est. monthly cost |
|---|---|---|---|
| Light chatbot | 10,000 | 800 / 400 | $2.60 |
| Coding agent | 5,000 | 8,000 / 2,000 | $8.00 |
| RAG pipeline | 30,000 | 4,000 / 500 | $16.50 |
| Doc summarizer | 2,000 | 20,000 / 1,000 | $4.00 |
Estimates use list prices without prompt caching. Caching can cut input cost by up to 90% on repeated context.
About PrismML: Ternary Bonsai 2 27B
Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and image understanding with a 262K-token context window. Ternary compression shrinks...
Compare PrismML: Ternary Bonsai 2 27B with
- 2 27B vs OpenAI: GPT-5.6 Terra Pro
- 2 27B vs Anthropic: Claude Fable 5
- 2 27B vs Anthropic: Claude Opus 5
- 2 27B vs OpenAI: GPT-5.6 Sol
Popular head-to-heads — see all on the comparisons index.
Frequently asked questions
How much does PrismML: Ternary Bonsai 2 27B cost?
PrismML: Ternary Bonsai 2 27B costs $0.075 per 1M input tokens and $0.500 per 1M output tokens; cached input tokens cost $0.037 per 1M. Verified 2026-10-21 — provider source.
What is the context window?
262K tokens, with up to 33K output tokens per response. That's roughly 328 pages of text in a single request.
Does it support prompt caching?
Yes — cached input is $0.037/1M, about 50% cheaper than standard input. Agents and RAG workloads with repeated context benefit most.
What's a cheaper alternative?
ScaleDown: Summarize saves about 85% on a typical workload with comparable quality.
Official resources
- PrismML: Ternary Bonsai 2 27B on OpenRouter — live endpoint, pricing source of record.
- Artificial Analysis leaderboard → — independent intelligence and performance rankings.
- Shred your current bill — paste a CSV and see if switching saves money.