- Rank
- #232
- Total Output Tokens
- 1,016,785
- Response Time (avg)
- 85.44s
- Total Cost
- $0.756
Trinity Large Thinking (medium) vs Gemini 2.5 Flash
Trinity Large Thinking (medium) leads on average score with 6.1 vs 6.0. Gemini 2.5 Flash has the lower benchmark cost at $0.017 vs $0.756. Gemini 2.5 Flash is faster at 6.19s vs 85.44s, with pass rates of 48.5% vs 39.4%.
Compared models
- Rank
- #238
- Total Output Tokens
- 1,890
- Response Time (avg)
- 6.19s
- Total Cost
- $0.017
Recommended model
Gemini 2.5 Flash
Its score stays close to the best score here (6.0 vs 6.1), while costing about 45.3x less than Trinity Large Thinking (medium).
Detailed comparison
| Metric | Trinity Large Thinking Trinity Large Thinking medium | Gemini 2.5 Flash Gemini 2.5 Flash none |
|---|---|---|
| Score | 6.1 | 6.0 |
| Rank | #232 | #238 |
| Reliability | 10.0 | 10.0 |
| Consistency | 7.9 | 9.6 |
| Attempts | 66/66 | 66/66 |
| Tests Correct | ||
| Attempt pass rate | 48.5% | 39.4% |
| Flaky tests | 6 | 1 |
| Total Runs | 66 | 66 |
| Cost per result | 9.972 | 0.209 |
| Total Cost | $0.756 | $0.017 |
| Input Price | $0.250 / 1M | $0.300 / 1M |
| Output Price | $0.800 / 1M | $2.500 / 1M |
| Total Input Tokens | 118,486 | 39,886 |
| Output Tokens | 156,166 | 1,890 |
| Reasoning Tokens | 860,619 | 0 |
| Response Time (avg) | 85.44s | 6.19s |
| Response Time (max) | 525.14s | 118.00s |
| Response Time (total) | 1879.75s | 136.23s |
| Parameters | 398B total (13B active) | ~350B total (~20B active) |
| Availability | Weights available | Closed |
Model generation showcase
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#232 Trinity Large Thinking
medium- Cost
- $0.016
- Time
- 75.9s
- Tokens
- 19,401 tok
#238 Gemini 2.5 Flash
none
Invalid SVG
- Cost
- $0.164
- Time
- 215.5s
- Tokens
- 65,659 tok
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Category Breakdown
| Anti-AI Tricks | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 6.8 | 9.8 | 50.0% | 0 | 5.11s | 663 | 1,531 | 6,078 | |
| Gemini 2.5 Flash | 3.0 | 10.0 | 0.0% | 0 | 582ms | 492 | 102 | 0 |
| Coding | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 7.5 | 10.0 | 66.7% | 0 | 179.02s | 7,248 | 7,413 | 351,155 | |
| Gemini 2.5 Flash | 5.5 | 10.0 | 33.3% | 0 | 736ms | 8,122 | 483 | 0 |
| Combined | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 2.9 | 6.0 | 16.7% | 1 | 382.70s | 93,292 | 35,814 | 269,745 | |
| Gemini 2.5 Flash | 3.0 | 10.0 | 0.0% | 0 | 61.19s | 16,470 | 486 | 0 |
| Data parsing and extraction | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 6.3 | 5.8 | 66.7% | 1 | 16.38s | 6,906 | 475 | 11,153 | |
| Gemini 2.5 Flash | 10.0 | 10.0 | 100.0% | 0 | 652ms | 7,257 | 279 | 0 |
| Domain specific | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 5.3 | 7.2 | 44.4% | 1 | 124.17s | 756 | 99,334 | 172,018 | |
| Gemini 2.5 Flash | 3.6 | 7.2 | 22.2% | 1 | 449ms | 642 | 12 | 0 |
| General Intelligence | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 10.0 | 10.0 | 100.0% | 0 | 32.22s | 501 | 7,035 | 7,774 | |
| Gemini 2.5 Flash | 5.0 | 10.0 | 0.0% | 0 | 615ms | 486 | 78 | 0 |
| Instructions following | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 4.4 | 6.9 | 16.7% | 1 | 3.99s | 684 | 77 | 3,798 | |
| Gemini 2.5 Flash | 10.0 | 10.0 | 100.0% | 0 | 590ms | 615 | 72 | 0 |
| Puzzle Solving | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 5.2 | 5.4 | 44.5% | 2 | 2.90s | 678 | 2,508 | 3,231 | |
| Gemini 2.5 Flash | 7.7 | 10.0 | 66.7% | 0 | 604ms | 558 | 132 | 0 |
| Tool Calling | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 10.0 | 10.0 | 100.0% | 0 | 5.26s | 7,551 | 299 | 1,372 | |
| Gemini 2.5 Flash | 10.0 | 10.0 | 100.0% | 0 | 1.91s | 5,088 | 234 | 0 |
| Trivia | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Trinity Large Thinking | 3.0 | 10.0 | 0.0% | 0 | 97.44s | 207 | 1,680 | 34,295 | |
| Gemini 2.5 Flash | 3.0 | 10.0 | 0.0% | 0 | 1.15s | 156 | 12 | 0 |
Quick Compare
Switch Comparison Pair
Trinity Large ThinkingmediumvsQwen3.6 FlashnoneTrinity Large ThinkingmediumvsGemma 4 31BnoneFree AvailableTrinity Large ThinkingmediumvsInklinglowTrinity Large ThinkinglowvsGemini 2.5 FlashnoneTrinity Large ThinkingmediumvsGemini 3.1 Flash LitenoneTrinity Large ThinkingmediumvsQwen3.7 FlashnoneTrinity Large ThinkingmediumvsQwen3.5 Plus 2026-04-20noneTrinity Large ThinkingmediumvsGemini 3.1 Flash LiteminimalTrinity Large ThinkingmediumvsNemotron 3 UltranoneFree AvailableClaude Sonnet 5nonevsTrinity Large ThinkingmediumTrinity Large ThinkingmediumvsGPT-5.6 LunalowGemini 2.5 FlashnonevsGPT-5 Nanomedium