Qwen3 Coder 480B A35B qwen3-coder-480b-a35b
qwen · qwen coding flagship
#19 overall #4 coding #2 agentic
- params
- 480B (A35B)
- arch
- moe
- context
- 1024k
- license
- apache-2.0 open weights
- released
- Jul 2025
- train compute
- 1.6e+24 FLOP
- downloads/30d
- 30.5k
# why ranked
Overall: #19 — score 18.0 ● low — a single signal; treat with caution
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.50 | 18.0 |
| arena_elo | 0.30 | — |
| bench_composite | 0.20 | — |
missing signals are dropped and the remaining weights renormalized — never imputed.
Coding: #4 — score 92.3 ● low — a single signal; treat with caution
| signal | weight | input (0–100) |
|---|---|---|
| aa_coding | 0.25 | — |
| aider_polyglot | 0.25 | — |
| arena_elo_code | 0.15 | — |
| swe_bench_verified | 0.35 | 92.3 |
missing signals are dropped and the remaining weights renormalized — never imputed.
Agentic: #2 — score 92.3 ● low — a single signal; treat with caution
| signal | weight | input (0–100) |
|---|---|---|
| swe_bench_verified | 1.00 | 92.3 |
# trend
# benchmarks
| benchmark | score | source | date |
|---|---|---|---|
| AA Intelligence Index | 18.0 | Artificial Analysis | — |
| AA Math Index | 39.3 | Artificial Analysis | — |
| SWE-bench Verified | 69.6 | SWE-bench | Aug 2025 |
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M blend | tps | uptime |
|---|---|---|---|---|---|---|---|
| DeepInfra | fp4 | 256k | $0.30 | $1.00 | $0.47 | — | 99.9% |
| unknown | 256k | $0.22 | $1.80 | $0.61 | — | 13.0% | |
| Venice | fp8 | 250k | $0.35 | $1.50 | $0.64 | — | 99.3% |
| Novita | fp8 | 256k | $0.38 | $1.55 | $0.67 | — | 100.0% |
| WandB | bf16 | 256k | $1.00 | $1.50 | $1.12 | — | 100.0% |
| Alibaba | unknown | 256k | $0.97 | $4.88 | $1.95 | — | 100.0% |
blended = (3·input + output) / 4 per 1M tokens · cheapest first
source aliases
- aa
qwen3-coder-480b,qwen3-coder-480b-a35b-instruct- arena
Qwen3-Coder-480B-A35B-Instruct- openrouter
qwen/qwen3-coder- swebench
Qwen3-Coder-480B-A35B-Instruct