VigilSAR-Bench — Procurement Dossiers
claude-code-cli:claude-fable-5
—
65.9% · ECE 0.295 · Safety Risk 91 · 3 critical
claude-code-cli:claude-fable-5-early-july
—
65.4% · ECE 0.34 · Safety Risk 283 · 11 critical
kimi-cli:kimi-k3
—
64.7% · ECE 0.298 · Safety Risk 244 · 9 critical
codex-cli:gpt-5.5-medium
—
62.2% · ECE 0.312 · Safety Risk 250 · 10 critical
claude-code-cli:claude-opus-5-high
—
61.7% · ECE 0.316 · Safety Risk 160 · 5 critical
codex-cli:gpt-5.5
—
61.7% · ECE 0.379 · Safety Risk 283 · 11 critical
codex-cli:gpt-5.4-mini
—
60.2% · ECE 0.335 · Safety Risk 116 · 4 critical
codex-cli:gpt-5.6-terra
—
58.6% · ECE 0.296 · Safety Risk 183 · 7 critical
codex-cli:gpt-5.6-sol
—
57.9% · ECE 0.313 · Safety Risk 233 · 9 critical
codex-cli:gpt-5.6-terra-xhigh
—
57.7% · ECE 0.328 · Safety Risk 158 · 6 critical
codex-cli:gpt-5.3-codex-spark
—
57.2% · ECE 0.316 · Safety Risk 225 · 9 critical
nvidia-build:thinkingmachines/inkling@default
—
56.0% · ECE 0.362 · Safety Risk 116 · 4 critical
codex-cli:gpt-5.6-luna
—
55.5% · ECE 0.303 · Safety Risk 258 · 10 critical
nvidia-build:z-ai/glm-5.2@default
—
54.5% · ECE 0.33 · Safety Risk 266 · 10 critical
nvidia-build:nvidia/nemotron-3-ultra-550b-a55b@default
—
54.4% · ECE 0.308 · Safety Risk 233 · 9 critical
nvidia-build:minimaxai/minimax-m3@default
—
53.6% · ECE 0.299 · Safety Risk 233 · 9 critical
gemini-cli:gemini-3.1-pro-preview
—
53.0% · ECE 0.415 · Safety Risk 383 · 15 critical
mlx-openai:qwen/qwen3.6-35b-a3b
—
52.8% · ECE 0.307 · Safety Risk 100 · 4 critical
nvidia-build:deepseek-ai/deepseek-v4-flash@default
—
52.5% · ECE 0.291 · Safety Risk 341 · 13 critical
gemini-cli:gemini-3.1-flash-lite
—
51.1% · ECE 0.368 · Safety Risk 400 · 16 critical
nvidia-build:qwen/qwen3.5-397b-a17b@default
—
49.0% · ECE 0.374 · Safety Risk 33 · 1 critical
nvidia-build:qwen/qwen3-next-80b-a3b-instruct@default
—
48.1% · ECE 0.369 · Safety Risk 183 · 7 critical
nvidia-build:mistralai/mistral-medium-3.5-128b@default
—
47.5% · ECE 0.332 · Safety Risk 425 · 17 critical
nvidia-build:google/gemma-4-31b-it@default
—
47.1% · ECE 0.299 · Safety Risk 100 · 4 critical
nvidia-build:nvidia/nemotron-3-nano-omni-30b-a3b-reasoning@default
—
46.9% · ECE 0.327 · Safety Risk 500 · 20 critical
nvidia-build:openai/gpt-oss-20b@default
—
46.7% · ECE 0.359 · Safety Risk 316 · 12 critical
nvidia-build:mistralai/mistral-nemotron@default
—
46.6% · ECE 0.372 · Safety Risk 300 · 12 critical
nvidia-build:nvidia/nemotron-3-super-120b-a12b@default
—
46.3% · ECE 0.365 · Safety Risk 333 · 13 critical
nvidia-build:nvidia/llama-3.3-nemotron-super-49b-v1.5@default
—
46.1% · ECE 0.319 · Safety Risk 408 · 16 critical
nvidia-build:meta/llama-3.3-70b-instruct@default
—
43.7% · ECE 0.326 · Safety Risk 75 · 3 critical
nvidia-build:mistralai/mistral-large-3-675b-instruct-2512@default
—
43.6% · ECE 0.304 · Safety Risk 350 · 14 critical
nvidia-build:mistralai/mistral-small-4-119b-2603@default
—
43.4% · ECE 0.403 · Safety Risk 625 · 25 critical
nvidia-build:nvidia/nemotron-3-nano-30b-a3b@default
—
43.2% · ECE 0.407 · Safety Risk 383 · 15 critical
nvidia-build:google/diffusiongemma-26b-a4b-it@default
—
42.7% · ECE 0.416 · Safety Risk 308 · 12 critical
nvidia-build:nvidia/ising-calibration-1.5-31b@default
—
42.0% · ECE 0.398 · Safety Risk 583 · 23 critical
nvidia-build:openai/gpt-oss-120b@high
—
41.6% · ECE 0.312 · Safety Risk 258 · 10 critical
nvidia-build:nvidia/nvidia-nemotron-nano-9b-v2@default
—
41.6% · ECE 0.371 · Safety Risk 608 · 24 critical
nvidia-build:meta/llama-4-maverick-17b-128e-instruct@default
—
41.1% · ECE 0.383 · Safety Risk 208 · 8 critical
nvidia-build:poolside/laguna-xs-2.1@default
—
40.9% · ECE 0.399 · Safety Risk 500 · 20 critical