Model profile
Evidence summary
Safety. Qwen3-Coder 480B A35B Instruct has an estimated Safety rank of #195; its 90% source-sensitivity interval is #23–#308. Its behavior-only rank is #188; company governance moves the combined estimate to #195. Published Safety evidence spans 4 eval lineages and 2 of 7 components. Its strongest relative result is AA-Omniscience (hallucination_rate, #66 of 345); its weakest is Arena Factuality — Text Arena (factuality-only weighting) (factuality_bt_rating, #88 of 111).
Freedom. Qwen3-Coder 480B A35B Instruct has an estimated Freedom rank of #379; its 90% source-sensitivity interval is #209–#558. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Cisco AI Defense Rolling Single-Turn Leaderboard (single_turn_attack_success_rate, #47 of 104); its weakest is Cisco AI Defense Rolling Single-Turn Leaderboard (single_turn_attack_success_rate, #47 of 104).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗qwen3-coder-480b-a35b-instruct
- Hugging Face ↗Qwen/Qwen3-Coder-480B-A35B-Instruct
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗repository created at proxy
Safety evals
Rank and direction are specific to the Safety portfolio. Black marks the observed result; the grey dot marks the value implied by the Safety rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #66 / 345 | ↓0.4493 | Source ↗official | |
| Arena Factuality — Text Arena (factuality-only weighting)factuality_bt_rating | #88 / 111 | ↑1417.0 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #74 / 117 | ↑0.2 | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #58 / 104 | ↓40.78 | Source ↗official |
Freedom evals
Rank and direction are specific to the Freedom portfolio. Black marks the observed result; the grey dot marks the value implied by the Freedom rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #47 / 104 | ↑40.78 | Source ↗official |