GPT-4.1 nano vs Ternary Bonsai 2 27B
Prices checked
GPT-4.1 nano takes 4 of 5 rounds against Ternary Bonsai 2 27B. It is also the cheaper of the two to run, at $0.16 per million tokens against $0.18.
OpenAIWins 4 roundsGPT-4.1 nano
1M context · tool calling · reads images
$0.16/M tokensCheapest price, at Poe · 7.6¢ per 1,000 pages
Ternary Bonsai 2 27B
262k context · tool calling · reads images
$0.18/M tokensOpenRouter's own price · 7.5¢ per 1,000 pages
Maker's price in (lower wins)
$0.10/M
$0.07/M
Maker's price out (lower wins)
$0.40/M
$0.50/M
Cheapest price (lower wins)
$0.16/M at Poe
none below the maker
Maker's price, blended (lower wins)
$0.18/M
$0.18/M
Cost per 1,000 pages (lower wins)
7.6¢
7.5¢
Context
1M tokens
262k tokens
Open weights
no
no
Tool calling
yes
yes
Structured output
yes
yes
Reads images
yes
yes
Reasoning
no
yes
Released
14 Apr 2025
18 Sept 2026
Providers selling it
24
6
Price we measure against
the maker's own
OpenRouter
Trains on your prompts (no is better)
not published
no provider to check
Providers selling both · 4
Kilo Gateway$0.17 / $0.18
NanoGPT$0.17 / $0.18
Nous Portal$0.17 / $0.18
OpenRouter$0.17 / $0.18
GPT-4.1 nano, then Ternary Bonsai 2 27B, per million tokens.
Go further
Add a third or fourth model, or change the weighting, in the tool.
Frequently asked
- Which is cheaper, GPT-4.1 nano or Ternary Bonsai 2 27B?
- GPT-4.1 nano, at $0.16 per million tokens against $0.18 for Ternary Bonsai 2 27B, blended 3 in : 1 out, read on 31 Oct 2026.
- Where is each one cheapest?
- GPT-4.1 nano: $0.16 at Poe. Ternary Bonsai 2 27B: nobody sells it below OpenRouter's own price with a stated reason, so $0.18 is the number to beat.
- Can I buy both from one provider?
- Yes. 4 providers sell both: Kilo Gateway, NanoGPT, Nous Portal, OpenRouter. The cheapest of them for the two together is Kilo Gateway.
- Which holds more context?
- GPT-4.1 nano, at 1M tokens against 262k.
Prices exclude tax, read 31 Oct 2026 from each provider's own catalog. They are per M tokens, blended 3 in : 1 out, and a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.
Related comparisons