Model profile
Qwen3 Coder 480B A35B Instruct
Alibabadeveloper
2025-07-22release date
#90 / 267overall rank
3eval lineages
Evidence summary
Qwen3 Coder 480B A35B Instruct has an estimated overall rank of #90; its 90% source-sensitivity interval is #15–#207. Its behavior-only rank is #79; company governance moves the combined estimate to #90. Published evidence spans 3 evals and 2 of 7 behavior components. Its strongest relative result is AA-Omniscience (hallucination_rate, #39 of 311); its weakest is BullshitBench v2 (clear_pushback_rate, #62 of 105).
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
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #39 / 311 | 0.4329 | ↓ lower | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #62 / 105 | 0.2 | ↑ higher | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #59 / 105 | 40.78 | ↓ lower | Source ↗official |