Qwen3.5 122B A10B qwen3.5-122b-a10b
qwen · qwen hybrid-reasoning
#37 overall
- params
- 125B (A10B)
- arch
- moe
- context
- 256k
- license
- apache-2.0 open weights
- released
- Feb 2026
- reasoning
- yes
- downloads/30d
- 818.6k
# architecture
- layers
- 48
- d_model
- 3072
- heads
- 32q / 2kv
- head dim
- 256
- experts
- top-8 of 256
- vocab
- 248320
- family
- qwen3_5_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: #37 — score 63.0 ● high — all signals present
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.70 | 58.5 |
| bench_composite | 0.30 | 73.3 |
benchmark panel evidence:
| complete panel | score (0–100) |
|---|---|
| aa-coding-v1 | 71.9 |
| browsecomp-v1 | 74.8 |
Coding: provisional — score 61.8 ● 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 | 61.8 |
| aider_polyglot | 0.30 | — |
| 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 Coding Index | 45.7 | Artificial Analysis | — |
| AA Intelligence Index | 17.7 | Artificial Analysis | — |
| Aa-lcr | 0.7 / 1 | LLM Stats | — |
| Ai2d | 0.9 / 1 | LLM Stats | — |
| Androidworld-sr | 0.7 / 1 | LLM Stats | — |
| Babyvision | 0.4 / 1 | LLM Stats | — |
| Bfcl-v4 | 0.7 / 1 | LLM Stats | — |
| Browsecomp | 0.6 / 1 | LLM Stats | — |
| Browsecomp-zh | 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 | — |
| Codeforces | 0.9 / 3000 | LLM Stats | — |
| Countbench | 1.0 | LLM Stats | — |
| Vlmsareblind | 1.0 / 1 | LLM Stats | — |
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M cache | price src | tps | uptime | ✓ |
|---|---|---|---|---|---|---|---|---|---|
| Aihubmix | 256k | $0.11 | $0.90 | — | via litellm | — | — | ||
| Libertai | 256k | $0.25 | $1.75 | — | via litellm | — | — | ||
| Alibaba | unknown | 256k | $0.26 | $2.08 | — | via openrouter | — | 100.0% | |
| SiliconFlow | fp8 | 256k | $0.26 | $2.08 | — | openrouter | — | 82.5% | ✓ |
| DeepInfra | 256k | $0.29 | $2.40 | — | via hfrouter | 62 | — | ||
| AtlasCloud | fp8 | 256k | $0.30 | $2.40 | — | atlascloud | — | 91.3% | ✓ |
| Novita | bf16 | 256k | $0.40 | $3.20 | — | novita | — | 100.0% | ✓ |
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
qwen3.5-122b-a10b- epoch
Qwen3.5-122B-A10B- hf
nvidia/Qwen3.5-122B-A10B-NVFP4- openrouter
qwen/qwen3.5-122b-a10b