Triple Validation Testing
Configure API endpoints and keys for different model providers: OpenAI, Anthropic, Gemini, Groq, Ollama, Base44 and more
Models
0
Tests Run
45
Avg Accuracy
65.9%
Avg Drift
-0.5%
Regressing
Best Model
neural-chat
96% accuracy
Most Efficient
1073.6
tinyllama
| Rank | Status | Model | Provider | GDPR | Security | Accuracy | Correct/Total | Tokens | Cost | Speed | Perf Score | Drift | Energy | Cost/Q | Errors | Endpoint |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
neural-chat:latest 3919 MB | ollama | Yes | Secure | 95.9% | 677/706 | 300,754426/q | $3.0075 | 0.11s/query | 846.0 | -0.5% | 215.9 | $0.00426/q$3.0075 total | ✓ | http://localhost:11434 | ||
mistral:7b 4170 MB | ollama | Yes | Secure | 89.5% | 632/706 | 291,576413/q | $2.9158 | 0.09s/query | 1026.8 | +1.3% | 246.2 | $0.00413/q$2.9158 total | ✓ | http://localhost:11434 | ||
zephyr:latest 3919 MB | ollama | Yes | Secure | 89.4% | 631/706 | 302,872429/q | $3.0287 | 0.12s/query | 764.9 | +1.6% | 195.2 | $0.00429/q$3.0287 total | ✓ | http://localhost:11434 | ||
4 | mistral:latest 4170 MB | ollama | Yes | Secure | 89.1% | 629/706 | 291,576413/q | $2.9158 | 0.10s/query | 932.5 | -1.4% | 223.6 | $0.00413/q$2.9158 total | ✓ | http://localhost:11434 | |
5 | llama3:latest 4445 MB | ollama | Yes | Secure | 88.1% | 622/706 | 250,057354/q | $2.5006 | 0.22s/query | 409.1 | +1.5% | 92.0 | $0.00354/q$2.5006 total | ✓ | http://localhost:11434 | |
6 | llava-phi3:3.8b 2791 MB | ollama | Yes | Secure | 87.5% | 618/706 | 299,110424/q | $2.9911 | 0.07s/query | 1194.9 | +1.8% | 428.1 | $0.00424/q$2.9911 total | ✓ | http://localhost:11434 | |
7 | llava-phi3:latest 2791 MB | ollama | Yes | Secure | 87.5% | 618/706 | 299,110424/q | $2.9911 | 0.08s/query | 1056.8 | -1.9% | 378.6 | $0.00424/q$2.9911 total | ✓ | http://localhost:11434 | |
8 | llava:7b 4514 MB | ollama | Yes | Secure | 87.0% | 614/706 | 295,106418/q | $2.9511 | 0.12s/query | 731.9 | -0.7% | 162.1 | $0.00418/q$2.9511 total | ✓ | http://localhost:11434 | |
9 | qwen2.5-coder:7b 4466 MB | ollama | Yes | Secure | 86.0% | 607/706 | 263,500373/q | $2.6350 | 0.20s/query | 431.1 | +4.9% | 96.5 | $0.00373/q$2.6350 total | ✓ | http://localhost:11434 | |
10 | bakllava:7b 4514 MB | ollama | Yes | Secure | 85.8% | 606/706 | 297,236421/q | $2.9724 | 0.10s/query | 899.0 | -2.9% | 199.1 | $0.00421/q$2.9724 total | ✓ | http://localhost:11434 | |
11 | qwen2.5-coder:3b 1841 MB | ollama | Yes | Secure | 85.1% | 601/706 | 263,500373/q | $2.6350 | 0.26s/query | 323.5 | -0.1% | 175.7 | $0.00373/q$2.6350 total | ✓ | http://localhost:11434 | |
12 | qwen2.5-coder:1.5b 940 MB | ollama | Yes | Secure | 83.0% | 586/706 | 263,500373/q | $2.6350 | 0.17s/query | 501.2 | -3.5% | 533.1 | $0.00373/q$2.6350 total | ✓ | http://localhost:11434 | |
13 | llama3.1:8b 4693 MB | ollama | Yes | Secure | 81.3% | 574/706 | 250,057354/q | $2.5006 | 0.21s/query | 386.4 | +3.7% | 82.3 | $0.00354/q$2.5006 total | ✓ | http://localhost:11434 | |
14 | llama3.1:latest 4693 MB | ollama | Yes | Secure | 81.3% | 574/706 | 250,057354/q | $2.5006 | 0.22s/query | 365.7 | +7.5% | 77.9 | $0.00354/q$2.5006 total | ✓ | http://localhost:11434 | |
15 | minicpm-v:latest 5220 MB | ollama | Yes | Secure | 81.0% | 572/706 | 248,674352/q | $2.4867 | 0.20s/query | 404.9 | -3.1% | 77.6 | $0.00352/q$2.4867 total | ✓ | http://localhost:11434 | |
16 | gemma3:1b 778 MB | ollama | Yes | Secure | 79.9% | 564/706 | 262,810372/q | $2.6281 | 0.38s/query | 211.2 | -1.6% | 271.5 | $0.00372/q$2.6281 total | ✓ | http://localhost:11434 | |
17 | qwen2.5-coder:32b 18932 MB | ollama | Yes | Secure | 79.5% | 561/706 | 263,500373/q | $2.6350 | 1.69s/query | 46.9 | +0.3% | 2.5 | $0.00373/q$2.6350 total | ✓ | http://localhost:11434 | |
18 | ManishThota/llava_next_video:latest 4487 MB | ollama | Yes | Secure | 78.6% | 555/706 | 299,816425/q | $2.9982 | 0.10s/query | 824.7 | -4.1% | 183.8 | $0.00425/q$2.9982 total | ✓ | http://localhost:11434 | |
19 | llava-llama3:8b 5289 MB | ollama | Yes | Secure | 77.2% | 545/706 | 250,057354/q | $2.5006 | 0.24s/query | 327.2 | +1.6% | 61.9 | $0.00354/q$2.5006 total | ✓ | http://localhost:11434 | |
20 | llava-llama3:latest 5289 MB | ollama | Yes | Secure | 77.2% | 545/706 | 250,057354/q | $2.5006 | 0.25s/query | 314.3 | +2.1% | 59.4 | $0.00354/q$2.5006 total | ✓ | http://localhost:11434 | |
21 | qwen2.5:3b 1841 MB | ollama | Yes | Secure | 74.8% | 528/706 | 263,500373/q | $2.6350 | 0.19s/query | 397.6 | -2.6% | 216.0 | $0.00373/q$2.6350 total | ✓ | http://localhost:11434 | |
22 | qwen2.5:14b 8572 MB | ollama | Yes | Secure | 74.2% | 524/706 | 263,500373/q | $2.6350 | 0.49s/query | 151.6 | -5.3% | 17.7 | $0.00373/q$2.6350 total | ✓ | http://localhost:11434 | |
23 | llava:13b 7640 MB | ollama | Yes | Secure | 67.6% | 477/706 | 299,816425/q | $2.9982 | 0.48s/query | 141.7 | +4.3% | 18.5 | $0.00425/q$2.9982 total | ✓ | http://localhost:11434 | |
24 | llama3.2:1b 1260 MB | ollama | Yes | Secure | 62.5% | 441/706 | 260,647369/q | $2.6065 | 0.19s/query | 328.7 | -1.2% | 260.9 | $0.00369/q$2.6065 total | ✓ | http://localhost:11434 | |
25 | qwen2.5-coder:14b 8572 MB | ollama | Yes | Secure | 60.2% | 425/706 | 263,500373/q | $2.6350 | 0.49s/query | 121.8 | -1.0% | 14.2 | $0.00373/q$2.6350 total | ✓ | http://localhost:11434 | |
26 | qwen2.5:0.5b 379 MB | ollama | Yes | Secure | 60.2% | 425/706 | 263,500373/q | $2.6350 | 0.17s/query | 355.1 | +1.7% | 936.9 | $0.00373/q$2.6350 total | ✓ | http://localhost:11434 | |
27 | qwen2.5-coder:0.5b 379 MB | ollama | Yes | Secure | 59.1% | 417/706 | 263,500373/q | $2.6350 | 0.16s/query | 372.6 | -1.6% | 983.1 | $0.00373/q$2.6350 total | ✓ | http://localhost:11434 | |
28 | tinyllama:1.1b 608 MB | ollama | Yes | Secure | 54.1% | 382/706 | 321,467455/q | $3.2147 | 0.08s/query | 652.8 | +4.1% | 1073.6 | $0.00455/q$3.2147 total | ✓ | http://localhost:11434 | |
29 | llama3.2:3b 1926 MB | ollama | Yes | Secure | 51.4% | 363/706 | 260,647369/q | $2.6065 | 0.28s/query | 183.2 | +1.2% | 95.1 | $0.00369/q$2.6065 total | ✓ | http://localhost:11434 | |
30 | llama3.2:latest 1926 MB | ollama | Yes | Secure | 51.4% | 363/706 | 260,647369/q | $2.6065 | 0.29s/query | 178.1 | +5.4% | 92.5 | $0.00369/q$2.6065 total | ✓ | http://localhost:11434 | |
31 | qwen3.8:27b 16920 MB | ollama | Yes | Secure | 51.3% | 362/706 | 270,214383/q | $2.7021 | 5.15s/query | 10.0 | -4.3% | 0.6 | $0.00383/q$2.7021 total | ✓ | http://localhost:11434 | |
32 | qwen3-vl:latest 5856 MB | ollama | Yes | Secure | 51.3% | 362/706 | 252,410358/q | $2.5241 | 0.30s/query | 173.3 | -13.4% | 29.6 | $0.00358/q$2.5241 total | ✓ | http://localhost:11434 | |
33 | glm4:latest 5203 MB | ollama | Yes | Secure | 51.3% | 362/706 | 245,818348/q | $2.4582 | 0.18s/query | 281.8 | -2.1% | 54.2 | $0.00348/q$2.4582 total | ✓ | http://localhost:11434 | |
34 | qwen3:latest 4983 MB | ollama | Yes | Secure | 51.3% | 362/706 | 250,086354/q | $2.5009 | 0.19s/query | 263.3 | +7.7% | 52.8 | $0.00354/q$2.5009 total | ✓ | http://localhost:11434 | |
35 | qwen3:8b 4983 MB | ollama | Yes | Secure | 51.3% | 362/706 | 250,086354/q | $2.5009 | 0.20s/query | 251.9 | -12.8% | 50.5 | $0.00354/q$2.5009 total | ✓ | http://localhost:11434 | |
36 | vicuna:latest 3649 MB | ollama | Yes | Secure | 51.3% | 362/706 | 320,292454/q | $3.2029 | 0.08s/query | 666.8 | -15.7% | 182.7 | $0.00454/q$3.2029 total | ✓ | http://localhost:11434 | |
37 | qwen3-vl:2b 1802 MB | ollama | Yes | Secure | 51.3% | 362/706 | 260,594369/q | $2.6059 | 0.28s/query | 182.9 | +8.6% | 101.5 | $0.00369/q$2.6059 total | ✓ | http://localhost:11434 | |
38 | moondream:latest 1658 MB | ollama | Yes | Secure | 51.3% | 362/706 | 306,818435/q | $3.0682 | 0.08s/query | 648.8 | +11.5% | 391.3 | $0.00435/q$3.0682 total | ✓ | http://localhost:11434 | |
39 | qwen3:1.7b 1296 MB | ollama | Yes | Secure | 51.3% | 362/706 | 250,086354/q | $2.5009 | 0.17s/query | 302.4 | -2.1% | 233.3 | $0.00354/q$2.5009 total | ✓ | http://localhost:11434 | |
40 | qwen3:0.6b 498 MB | ollama | Yes | Secure | 51.3% | 362/706 | 250,086354/q | $2.5009 | 0.17s/query | 309.1 | -11.0% | 620.6 | $0.00354/q$2.5009 total | ✓ | http://localhost:11434 | |
41 | llama2:latest 3650 MB | ollama | Yes | Secure | 49.3% | 348/706 | 306,170434/q | $3.0617 | 0.11s/query | 429.5 | -4.5% | 117.7 | $0.00434/q$3.0617 total | ✓ | http://localhost:11434 | |
42 | mapler/llava-mistral:latest 4766 MB | ollama | Yes | Secure | 48.7% | 344/706 | 307,872436/q | $3.0787 | 0.10s/query | 478.3 | -4.8% | 100.4 | $0.00436/q$3.0787 total | ✓ | http://localhost:11434 | |
43 | orca-mini:latest 1888 MB | ollama | Yes | Secure | 48.7% | 344/706 | 318,259451/q | $3.1826 | 0.08s/query | 604.0 | +10.7% | 319.9 | $0.00451/q$3.1826 total | ✓ | http://localhost:11434 |