Qwen3.6 27B qwen3.6-27b
qwen · qwen efficient
#15 overall
- params
- 27.8B
- arch
- dense
- context
- 256k
- license
- apache-2.0 open weights
- released
- Apr 2026
- reasoning
- yes
- downloads/30d
- 4.9M
# why ranked
Overall: #15 — score 39.1 ● medium — one signal missing, weights renormalized
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.50 | 37.1 |
| arena_elo | 0.30 | — |
| bench_composite | 0.20 | 44.2 |
missing signals are dropped and the remaining weights renormalized — never imputed.
Coding: provisional — score 72.4 ● low — a single signal; treat with caution
provisional: too few independent signals for a numbered position — the score is shown, but this model sorts after every ranked model.
| signal | weight | input (0–100) |
|---|---|---|
| aa_coding | 0.25 | 72.4 |
| aider_polyglot | 0.25 | — |
| arena_elo_code | 0.15 | — |
| swe_bench_verified | 0.35 | — |
missing signals are dropped and the remaining weights renormalized — never imputed.
# trend
# benchmarks
| benchmark | score | source | date |
|---|---|---|---|
| AA Coding Index | 53.7 | Artificial Analysis | — |
| AA Intelligence Index | 37.1 | Artificial Analysis | — |
| LiveBench Agentic Coding | 39.3 | LiveBench | Jun 2026 |
| LiveBench Coding | 71.8 | LiveBench | Jun 2026 |
| LiveBench Data Analysis | 70.4 | LiveBench | Jun 2026 |
| LiveBench IF | 53.2 | LiveBench | Jun 2026 |
| LiveBench Language | 63.3 | LiveBench | Jun 2026 |
| LiveBench Mathematics | 79.9 | LiveBench | Jun 2026 |
| LiveBench Reasoning | 70.3 | LiveBench | Jun 2026 |
| AIME 2026 | 0.9 / 1 | LLM Stats | — |
| Androidworld | 0.7 / 1 | LLM Stats | — |
| C-eval | 0.9 / 1 | LLM Stats | — |
| Cc-ocr | 0.8 / 1 | LLM Stats | — |
| Charxiv-r | 0.8 / 1 | LLM Stats | — |
| Claw-eval | 0.6 / 1 | LLM Stats | — |
| Countbench | 1.0 | LLM Stats | — |
| Dynamath | 0.9 / 1 | LLM Stats | — |
| Gdpval-aa | 1158.0 / 3000 | LLM Stats | — |
| Qwenwebbench | 1487.0 / 2000 | LLM Stats | — |
| Vlmsareblind | 1.0 / 1 | LLM Stats | — |
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M blend | tps | uptime |
|---|---|---|---|---|---|---|---|
| Chutes | fp8 | 256k | $0.30 | $2.00 | $0.72 | — | 95.9% |
| Io Net | fp8 | 262k | $0.28 | $2.40 | $0.81 | — | 96.3% |
| Morph | unknown | 128k | $0.29 | $2.40 | $0.82 | — | 86.6% |
| Phala | unknown | 256k | $0.32 | $2.70 | $0.92 | — | 98.1% |
| Alibaba | unknown | 256k | $0.45 | $2.70 | $1.01 | — | 99.7% |
| SiliconFlow | fp8 | 256k | $0.30 | $3.20 | $1.02 | — | 82.0% |
| DeepInfra | fp8 | 256k | $0.32 | $3.20 | $1.04 | — | 99.8% |
| Venice | fp8 | 250k | $0.33 | $3.25 | $1.06 | — | 99.4% |
| WandB | fp8 | 256k | $0.60 | $3.60 | $1.35 | — | 100.0% |
blended = (3·input + output) / 4 per 1M tokens · cheapest first
source aliases
- aa
qwen3.6-27b- openrouter
qwen/qwen3.6-27b