Qwen3-VL-235B-A22B-Thinking qwen3-vl-235b-a22b-thinking
qwen · qwen vision reasoning flagship
#29 overall
- params
- 235.7B (A22B)
- arch
- moe
- context
- 128k
- license
- apache-2.0 open weights
- reasoning
- yes
# architecture
- layers
- 94
- d_model
- 4096
- heads
- 64q / 4kv
- head dim
- 128
- experts
- top-8 of 128
- vocab
- 151936
- family
- qwen3_vl_moe_text
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: #29 — score 57.3 ● high — all signals present
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.70 | 49.5 |
| bench_composite | 0.30 | 75.6 |
benchmark panel evidence:
| complete panel | score (0–100) |
|---|---|
| aime-2025-v1 | 75.6 |
# trend
methodology changed during this history window; score movement across that boundary is not model movement. See methodology v5.
# benchmarks
| benchmark | score | source | date |
|---|---|---|---|
| AA Intelligence Index | 14.4 | Artificial Analysis | — |
| AA Math Index | 88.3 | Artificial Analysis | — |
| Ai2d | 0.9 / 1 | LLM Stats | — |
| AIME 2025 | 0.9 / 1 | LLM Stats | — |
| Bfcl-v3 | 0.7 / 1 | LLM Stats | — |
| Blink | 0.7 / 1 | LLM Stats | — |
| Cc-ocr | 0.8 / 1 | LLM Stats | — |
| Charadessta | 0.6 / 1 | LLM Stats | — |
| Charxiv-r | 0.7 / 1 | LLM Stats | — |
| Countbench | 0.9 | LLM Stats | — |
| Creative-writing-v3 | 0.9 / 1 | LLM Stats | — |
| Docvqatest | 1.0 / 1 | LLM Stats | — |
| Mm-mt-bench | 8.5 | LLM Stats | — |
| Screenspot | 1.0 / 1 | LLM Stats | — |
| Zebralogic | 1.0 / 1 | LLM Stats | — |
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M cache | price src | tps | uptime | ✓ |
|---|---|---|---|---|---|---|---|---|---|
| Featherless | — | — | — | — | via hfrouter | — | — | ||
| AtlasCloud | fp8 | 128k | $0.50 | $2.50 | — | atlascloud | — | — | ✓ |
| SiliconFlow | 262k | $0.45 | $3.50 | — | via modelsdev | — | — | ||
| Alibaba | unknown | 128k | $0.40 | $4.00 | — | via openrouter | — | 88.0% | |
| Qwen Ai Platform | 128k | $0.40 | $4.00 | — | via litellm | — | — | ||
| Qwencloud | 128k | $0.40 | $4.00 | — | via litellm | — | — | ||
| Novita | bf16 | 128k | $0.98 | $3.95 | — | novita | — | 99.8% | ✓ |
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
qwen3-vl-235b-a22b-reasoning- llmstats
qwen3-vl-235b-a22b-thinking- openrouter
qwen/qwen3-vl-235b-a22b-thinking