Qwen3.5-122B-A10B
Qwen · Open weights · Released Feb 25, 2026 · qwen/qwen3.5-122b-a10b
Best $0.26 in / $2.08 out via Alibaba
262K contextVisionToolsReasoningStructured outputPrompt cachingOpen weights
Available from 4 providersUSD / MTOK · PRICES 23 MINUTES AGO · SPEED 5 DAYS AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | SMOKE | UPTIME |
|---|---|---|---|---|---|---|---|---|---|
 Alibaba | $0.26 | $2.08 | — | 262K | — | 1.5s | 159 tok/s | 5/5 | 97.3% |
 SiliconFlow | $0.26 | $2.08 | — | 262K | fp8 | ✕ | ✕ | 2/5 | 95.4% |
 AtlasCloud | $0.30 | $2.40 | $0.30 | 262K | fp8 | 1.9s | 129 tok/s | 5/5 | 99.0% |
 Novita | $0.40 | $3.20 | — | 262K | bf16 | 1.8s | 128 tok/s | 4/5 | 99.8% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · ✕ = probed, endpoint returned no measurable stream · SMOKE: of 5 programmatic capability probes passed (JSON schema, instruction following, tool call, long-context, benign compliance; retry-once) · screened via OpenRouter routing · v0.2 · methodology v1.1 · measured 5 days ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Recent changesRSS ↗
Oct 2, 2026endpoint removed · deepinfra5 → 4
About
The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of...
Specifications
Context window262,144
Max output65,536
Modalities intext, image, video
Modalities outtext
LicenseOpen weights
ReleasedFeb 25, 2026
TokenizerQwen3
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/qwen/qwen3.5-122b-a10b)Sources
OpenRouter API23 minutes ago
AGGREGATORHugging Face23 minutes ago
COMMUNITYModelIndex benchmarks5 days ago
MEASUREDVerified 23 minutes ago · list prices, not negotiated rates.