AI Safety Evidence

← All models

Developer

Alibaba

108 indexed models; 57 currently meet the evidence threshold for the overall ranking. Together they have results from 99 evaluations.

Official Alibaba model information ↗

Company governance evidence

Alibaba is represented at -1.00 SD relative to the matched Future of Life Institute edition. Provenance: Direct. This company-level evidence contributes 10% of overall model rank.

Models by Alibaba

ModelEvalsComponentsRankRelease date
Qwen3.5 397B A17B137/7442026-02-16
Qwen3.5 27B76/7452026-02-24
Qwen3.5 9B57/7672026-02-27
Qwen3.5 Plus76/7732026-02-15
Qwen3.7 Flash32/7742026-07-27
Qwen3 235B A22B FP8 Throughput63/7762025-04-28
Qwen3.7 Max137/7802026-05-20
Qwen3.8 Max76/7892026-08-03
Qwen 72B Chat44/7952023-11-30
Qwen2 72B Instruct105/7962024-06-07
Qwen3 Max96/7992025-09-24
Qwen3 4B45/71022025-04-27
Qwen3.6 27B45/71072026-04-21
Qwen3 30B A3B65/71222025-04-27
Qwen2.5 72B Instruct197/71282024-09-16
Qwen3 235B A22B327/71322025-04-29
Qwen3 32B97/71452025-04-29
Qwen3.6 Plus117/71462026-04-02
Qwen 14B Chat84/71532023-09-25
Qwen3.6 Max Preview53/71542026-04-22
Qwen3.8 27B43/71552026-08-14
Qwen2.5 32B Instruct55/71582024-09-19
Qwen3.7 Plus75/71602026-05-26
Qwen3 Max Preview32/71612025-09-05
Qwen3 Next 80B A3B126/71652025-09-11
Qwen3 235B A22B Instruct34/71702025-07-21
Qwen3.5 Flash65/71752026-02-24
Qwen1.5 32B Chat46/71762024-04-02
Qwen1.5 14B Chat65/71832024-02-04
Qwen3 30B A3B Instruct76/71872025-07-28
Qwen3-Coder 480B A35B Instruct42/71952025-07-22
Qwen1.5 72B Chat136/71982024-02-04
Qwen1.5 7B Chat96/72062024-02-04
Qwen3.5 122B A10B55/72392026-02-24
Qwen3.5 4B35/72412026-02-27
Qwen 7B Chat94/72472023-08-03
Qwen3.5 35B A3B54/72492026-02-24
Qwen2.5 7B Instruct207/72522024-09-19
Qwen2.5 14B Instruct65/72582024-09-19
Qwen3-VL 32B Instruct33/72672025-10-19
Qwen2 7B Instruct54/72762024-06-07
Qwen 2.5 Max46/72792025-01-25
Qwen2.5 1.5B Instruct53/72822024-09-19
Qwen2.5 3B Instruct33/72832024-09-17
Qwen1.5 0.5B Chat34/72852024-02-04
Qwen3 14B66/72892025-04-27
SeaLLMs v3 7B Chat34/72912024-07-03
Qwen3 8B106/72922025-04-29
Qwen3 Next 80B A3B Instruct42/72952025-09-09
Qwen3-VL 235B A22B Instruct44/72972025-09-22
Qwen2.5 7B33/73022024-09-19
Qwen3-VL 8B Instruct44/73072025-10-11
Qwen3 4B Instruct32/73122025-04-29
QwQ 32B45/73152025-03-06
Qwen 1.8B Chat33/73202023-11-30
Qwen1.5 1.8B Chat45/73262024-02-04
Qwen2.5 0.5B Instruct43/73302024-09-19
ChatPLUG 13B11/7——
ChatPLUG 3.7B11/7—2023-04-19
Qwen Plus23/7——
Qwen Turbo11/7—2024-11-18
Qwen1.5 110B Chat24/7—2024-04-25
Qwen1.5 4B Chat24/7—2024-02-04
Qwen2 0.5B Instruct11/7—2024-06-03
Qwen2 1.5B Instruct21/7—2024-06-07
Qwen2 57B A14B Instruct13/7—2024-06-04
Qwen2 7B11/7—2024-06-04
Qwen2 7B Chat11/7—2024-06-07
Qwen2-VL 72B Instruct23/7—2024-09-17
Qwen2.5 14B Instruct 1M13/7—2025-01-23
Qwen2.5 7B Instruct 1M13/7—2025-01-23
Qwen2.5-Coder 0.5B Instruct11/7—2024-11-06
Qwen2.5-Coder 1.5B Instruct11/7—2024-09-18
Qwen2.5-Coder 3B Instruct11/7—2024-11-06
Qwen2.5-Math 1.5B Instruct11/7—2024-09-16
Qwen3 0.6B21/7—2025-04-27
Qwen3 0.6B Instruct11/7—2025-04-28
Qwen3 1.7B11/7—2025-04-27
Qwen3 1.7B Instruct11/7—2025-04-28
Qwen3 14B Instruct11/7—2025-04-28
Qwen3 32B Instruct11/7—2025-04-27
Qwen3 4B Thinking11/7—2025-08-05
Qwen3 8B Base11/7—2025-04-28
Qwen3 8B Instruct11/7—2025-04-28
Qwen3-Coder 30B A3B Instruct22/7—2025-07-31
Qwen3-Coder Next22/7—2026-01-30
Qwen3-Omni 30B A3B11/7—2025-09-22
Qwen3-Omni 30B A3B Instruct11/7—2025-09-20
Qwen3-VL 235B A22B22/7—2025-09-22
Qwen3-VL 30B A3B22/7—2025-09-30
Qwen3-VL 30B A3B Instruct22/7—2025-09-30
Qwen3-VL 32B22/7—2025-10-21
Qwen3-VL 4B11/7—2025-10-14
Qwen3-VL 4B Instruct11/7—2025-10-11
Qwen3-VL 8B22/7—2025-10-11
Qwen3.512/7—2026-02-16
Qwen3.5 0.8B11/7—2026-02-28
Qwen3.5 2B11/7—2026-02-28
Qwen3.5 Max11/7——
Qwen3.5-Omni Flash11/7—2026-03-30
Qwen3.5-Omni Plus11/7—2026-03-30
Qwen3.6 35B A3B11/7—2026-04-15
Qwen3.6 Flash00/7——
Qwen3.8 Flash11/7—2026-08-26
Qwen3.8 Flash Next11/7—2026-08-26
QwQ 32B Preview13/7—2024-11-27
SeaLLM 7B v213/7——
SeaLLMs v3 7B11/7——
Evaluations covering Alibaba models (99)

AA-Omniscience · AbstentionBench · Adversarial Humanities Benchmark (AHB) — Table 5 · Adversarial Poetry — AILuminate Baseline and Poetry ASR · Agent-SafetyBench · AgentDrive Safety Compliance · AILuminate General Purpose AI Chat · AIRBench 2024 Safety Scenarios · Alignment Leaderboard · Anthropic Agentic Misalignment — blackmail · Anthropic Agentic Misalignment — corporate espionage · Arena Factuality — Text Arena (factuality-only weighting) · AuAu Authoritarian Response Audit · BioTIER · BrokenMath · BullshitBench v2 · CAIS Risk Index · CASE-Bench · ChineseSafe · ChiSafetyBench · Cisco AI Defense Rolling Single-Turn Leaderboard · CMoralEval · COMPL-AI AI-Identity Disclosure · COMPL-AI LLM RuLES Multi-Turn Rule Following · COMPL-AI TensorTrust Goal-Hijacking Resistance · Concordia AI Risk Monitor · Confabulations · Contextual MoralChoice · CRiskEval · CValues · DelusionEval · DSPSafeBench · DystopiaBench · Emergent Collusion · Enkrypt AI Safety Leaderboard · Every Model Cheats — Cybench Cheat Propensity · Fake Alignment (FINE) · FinEval 6.0 Safety Awareness · FinEval Financial Security Knowledge · FlagEval Safety and Values · FLAMES · FORTRESS · Governance Decay under Passive Context Compaction · HalluVerse-M3 Hallucination Recognition · HarmBench · HarmVideoBench · HELM Safety · HUMAINE Trust, Ethics and Safety · Human Pathogen Capabilities Test (HPCT) — overall refusal · HyperCLOVA X Toxic Continuation Panels · IndoBias-Pairs — parity-aware culturally grounded bias · Inkling-Small model card — FORTRESS · Inkling-Small model card — StrongREJECT · JailBench · JuICE Cultural-Error Span Detection · KIDBench Implicit Child Cue · LiveSecBench · M3-SafetyBench · Manager Coercion Bench · MASK · MT-JailBench CrescendoX · MuPPET Contextual Privacy · NESSiE Necessary Safety Benchmark · ODCV-Bench · Open LLM Safety Index · Opposite-Narrator Sycophancy · OR-Bench · PacifAIst · PandaBench JBB direct-request panel · PHARE · Pokee-Isaac model card — DTAP · PropensityBench · Qwen2 Safety Panel · RefusalBench · S-Eval · SABER · SafeArena · SafeDialBench · SafetyBench · SALAD-Bench · Shell · SM-Bench · Social Welfare Function Benchmark · SORRY-Bench · SOSBench · SpeciesismBench · SpeciEval · StereoTales Harmful Associations · SuperCLUE Safety · SYCON Bench · TAC · ThaiSafetyBench · The Dictatorship Eval · ToolPrivacyBench · TrustLLM contemporary collapsed application · UAVBench safety-critical decision recognition · Vectara HHEM Factual Consistency · VETO Misfired Alignment · WildClawBench Safety & Alignment (OpenClaw harness)