Model Library
Trillion LabsReleased February 10, 2026Tri-21B-Think
Trillion Labs · Released February 10, 2026
Intelligence score
12
At a glance
- Intelligence
- 12
- Speed
- —
- Latency
- —
- Input price
- Not published
- Output price
- Not published
- Context window
- 32Ktokens
- Parameters
- 21billions
Summary
Comparison summary
Tri-21B-Think scores 12 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight models of similar size (median: 9).
Specifications
Technical specifications
| Model name | Tri-21B-Think |
|---|---|
| Provider | Trillion Labs |
| Release date | February 10, 2026 |
| Intelligence score | 12 |
| Speed | — |
| Latency (TTFT) | — |
| Input price | Not published |
| Output price | Not published |
| Context window | 32,000 tokens |
| Parameters | 21B |
| Reasoning | Yes |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | Apache 2 |
| API availability | — |
| Source | Model Intelligence legacy profile: tri-21b-think-v0-5 |
FAQ · 15
Frequently asked questions
Tri-21B-Think was released on February 10, 2026.
Tri-21B-Think was created by Trillion Labs.
Tri-21B-Think scores 12 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight models of similar size (median: 9).
Yes, Tri-21B-Think is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
Tri-21B-Think supports text input.
Tri-21B-Think supports text output.
No, Tri-21B-Think does not support image input. It can only process text.
No, Tri-21B-Think is not multimodal. It only supports text input.
Tri-21B-Think has a context window of 32k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Tri-21B-Think is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Tri-21B-Think has 21 billion parameters.
Tri-21B-Think is released under the Apache 2.0 license. This license allows commercial use.
Tri-21B-Think achieves a score of 12 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Tri-21B-Think is an open weights model that can be self-hosted.
Tri-21B-Think is an open weights model that can be downloaded and self-hosted.