QwQ 32B qwq-32b
qwen · qwen reasoning
#85 overall
- params
- 32.8B
- arch
- dense
- license
- apache-2.0 open weights
- released
- Mar 2025
- reasoning
- yes
- train compute
- 3.5e+24 FLOP
# architecture
- layers
- 64
- d_model
- 5120
- heads
- 40q / 8kv
- head dim
- 128
- vocab
- 152064
- family
- qwen2
signal path of one layer, generated from the registry's structured fields — dimension lines quote the model's real numbers; an MoE trace forks at the router, a dense trace runs straight through. Geometry fields come from the repo's config.json.
# why ranked
Overall: #85 — score 24.9 ● high — all signals present
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.70 | 28.9 |
| bench_composite | 0.30 | 15.8 |
benchmark panel evidence:
| complete panel | score (0–100) |
|---|---|
| aa-math-v1 | 5.6 |
| aider-polyglot-v1 | 26.0 |
Coding: provisional — score 27.3 ● 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 | 0.30 | — |
| aider_polyglot | 0.30 | 27.3 |
| swe_bench_verified | 0.40 | — |
missing signals are dropped and the remaining weights renormalized — never imputed.
# trend
methodology changed during this history window; score movement across that boundary is not model movement. See methodology v6.
# benchmarks
| benchmark | score | source | date |
|---|---|---|---|
| AA Intelligence Index | 9.5 | Artificial Analysis | — |
| AA Math Index | 29.0 | Artificial Analysis | — |
| Aider Polyglot | 20.9 | Aider | Mar 2025 |
| AIME 2024 | 0.8 / 1 | LLM Stats | — |
| Bfcl | 0.7 / 1 | LLM Stats | — |
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M cache | price src | tps | uptime | ✓ |
|---|---|---|---|---|---|---|---|---|---|
| Nscale | — | $0.18 | $0.20 | — | via litellm | — | — | ||
| Hyperbolic | 128k | $0.20 | $0.20 | — | via litellm | — | — | ||
| DeepInfra | 128k | $0.15 | $0.40 | — | via litellm | — | — | ||
| Nebius | 32k | $0.15 | $0.45 | — | via litellm | — | — | ||
| Cloudflare | 24k | $0.66 | $1.00 | — | via modelsdev | — | — | ||
| Together | 128k | $1.20 | $1.20 | — | via litellm | — | — |
sorted by blended price ((3·input + output) / 4 per 1M) · ✓ = the provider's own catalog confirms the offer · "via …" prices are what the aggregator routing the offer charges, not the provider's own list price
source aliases
- aa
qwq-32b- arena
QwQ-32B- openrouter
qwen/qwq-32b