- Score
- 8.6 (+2.84 points)
- Total Cost (Current Price)
- $0.061 · 92.4% cheaper
- Response Time (avg)
- 7.51s · 89.9% faster
Arcee AI
Release: 2026-07-28
Tested on: 2026-10-01 03:58
arcee-ai/trinity-large-thinking::high
398B total (13B active)MoEWeights available
Summary
Trinity Large Thinking scores 5.7 on AI BENCHY and ranks #310. It has 10.0 reliability, a 42.0% pass rate, $0.794 total cost, and 74.52s average response time.
What makes Trinity Large Thinking unique: It stands out most in Agentic, where it ranks #1, while Coding is its weakest area at #17. It uses unusually many reasoning tokens, which can help explain its slower or more expensive runs.
Recommended alternatives
We recommend choosing one of these models instead:
- Score
- 9.4 (+3.63 points)
- Total Cost (Current Price)
- $0.183 · 77.0% cheaper
- Response Time (avg)
- 6.24s · 91.6% faster
- Score
- 8.3 (+2.53 points)
- Total Cost (Current Price)
- $0.047 · 94.1% cheaper
- Response Time (avg)
- 18.07s · 75.8% faster
Model facts
Researched on 2026-08-12
- Parameters
- 398B total (13B active)
- Architecture
- MoE
- Availability
- Weights available
- License
- OpenMDW
5.7
Consistency
7.3
10.0
Total Output Tokens
717,762
Total Input Tokens
283,116
Input Price
$0.250 / 1M
Output Price
$0.800 / 1M
Cache Read Price
$0.060 / 1M
Cache Write Price
N/A
Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.
Flaky tests
8
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#310 Trinity Large Thinking
high- Cost
- $0.028
- Time
- 130.1s
- Tokens
- 34,387 tok
Run history
| Tested on | Score | Reliability | Tests Correct | Total Cost | Compare |
|---|---|---|---|---|---|
| 2026-10-01 03:58 Re-test | 5.7 | 10.0 | $0.794 ↑ | Current run | |
| 2026-07-28 10:47 Initial run | 4.8 | 9.9 | $0.632 | Compare |
This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.
Run comparison
| Run | Benchmark coverage | Score | Consistency | Reliability | Tests Correct | Flaky tests | Total Output Tokens | Total Input Tokens | Total Cost | Response Time (avg) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-10-01 03:58 · Current run | 69/69 attempts | 5.7 | 7.3 | 10.0 | 6/23 | 8 | 717,762 | 283,116 | $0.794 | 74.52s |
| 2026-07-28 10:47 · Initial run | 66/66 attempts | 4.8 | 7.2 | 9.9 | 5/22 | 8 | 953,758 | 101,767 | $0.632 | 77.22s |
| Difference | — | +0.9 | +0.1 | +0.1 | +1 | 0 | -235996 | +181349 | +$0.162 | -2705ms |
Benchmark coverage differs: 69/69 attempts (Target: 3 repeats per test) versus 66/66 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.
These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.
Price History
Historical pricing data for this model from OpenRouter.
| Date | Input Price | Output Price | Cache Read Price | Cache Write Price |
|---|---|---|---|---|
| 2026-08-14 12:51 | $0.220 / 1M | $0.850 / 1M | N/A | N/A |
| 2026-10-01 18:33 | $0.250 / 1M | $0.800 / 1M | $0.060 / 1M | N/A |
Charts
Choose the first model, then click a second model to open a side-by-side page.
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Quick Compare
Trinity Large ThinkinghighvsRing 2.6 1tmediumTrinity Large ThinkinghighvsGPT-5 NanomediumTrinity Large ThinkinghighvsLing 3.0 FlashhighTrinity Large ThinkinghighvsGPT-5.4 MininoneTrinity Large ThinkinghighvsDeepSeek V4.1 FlashnoneTrinity Large ThinkinghighvsMistral Large 4noneTrinity Large ThinkinghighvsGPT-5.6 TerranoneTrinity Large ThinkinghighvsSeed 2.1 TurbononeTrinity Large ThinkinghighvsDeepSeek V4.1 Flash 1017noneTrinity Large ThinkinghighvsDots 3 Note PreviewmediumFree Available
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Agentic | 10.0 | 10.0 | |
| Anti-AI Tricks | 6.7 | 8.1 | |
| Coding | 3.7 | 4.7 | |
| Combined | 3.0 | 10.0 | |
| Data parsing and extraction | 6.3 | 5.8 | |
| Domain specific | 2.9 | 7.2 | |
| General Intelligence | 6.5 | 3.4 | |
| Instructions following | 6.3 | 10.0 | |
| Puzzle Solving | 4.7 | 5.2 | |
| Tool Calling | 10.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |