AI model
qwen/qwen3.6-35b-a3b
Qwen: Qwen3.6 35B A3B is a 262,144-token model served on Composite at $0.0137 per million tokens in and $0.126 per million tokens out. Ranked #72 of 179 models by traffic on Composite over the last 30 days, taking 0.05% of tokens and 0.03% of requests. Measured on Composite, 92.1% of its last 89 requests on Composite succeeded, the typical first token arrives in 39101 ms, output runs at about 140.2 tokens per second.
Try in Playground| Model ID | qwen/qwen3.6-35b-a3b |
|---|---|
| Provider | cv11 |
| Context window | 262,144 tokens |
| Accepts | text, image, video |
| Returns | text |
| Input price | $0.0137 per million tokens |
| Output price | $0.126 per million tokens |
| Prompt caching | Supported — cached input is billed at the provider's cache rate, on credits and plans alike |
| Moderated | No |
| Cost on a monthly plan | About 0.2 plan credits per typical call (8k in, 1k out) |
| Routing endpoints | 10 (base plus 9 routing variants) |
Make a Composite API key and type qwen/qwen3.6-35b-a3b as the model name. JanitorAI takes the full URL https://composite.lucidity.sh/v1/chat/completions; SillyTavern, ChubAI, Agnai and RisuAI take https://composite.lucidity.sh/v1. It runs on credits, a plan, or your own provider key.
JanitorAI setup · SillyTavern setup · Best Janitor AI proxies compared
#72 of 179
0.05% of tokens and 0.03% of requests over 30 days.
92.1%
Across the last 89 requests Composite served for this model.
39101 ms
Median across the same sample.
140.2 tok/s
Median generation rate after the first token.
8.9%
Share of prompt tokens served from the provider's cache over 7 days.
The same model behind 9 alternative endpoints. Call the id directly to pin a route; the base id above picks for you.
| Route | Model ID | Provider | Price |
|---|---|---|---|
| Normal | qwen/qwen3.6-35b-a3b-normal |
cv11 | $0.125 per million tokens input · $1.13 per million tokens output |
| Normal + cheapThink | qwen/qwen3.6-35b-a3b-normal:cheapThink |
cv11 | $0.125 per million tokens input · $1.13 per million tokens output |
| Cheapest provider | qwen/qwen3.6-35b-a3b:cheapest-provider |
PrimeIntellect | $0.0137 per million tokens input · $0.126 per million tokens output |
| Cheapest provider + cheapThink | qwen/qwen3.6-35b-a3b:cheapest-provider:cheapThink |
PrimeIntellect | $0.0137 per million tokens input · $0.126 per million tokens output |
| cheapThink | qwen/qwen3.6-35b-a3b:cheapThink |
cv11 | $0.0137 per million tokens input · $0.126 per million tokens output |
| Fast | qwen/qwen3.6-35b-a3b:fast |
PrimeIntellect | $0.0137 per million tokens input · $0.126 per million tokens output |
| Fast + cheapThink | qwen/qwen3.6-35b-a3b:fast:cheapThink |
PrimeIntellect | $0.0137 per million tokens input · $0.126 per million tokens output |
| Quality | qwen/qwen3.6-35b-a3b:quality |
CoreWeave | $0.125 per million tokens input · $1.11 per million tokens output |
| Quality + cheapThink | qwen/qwen3.6-35b-a3b:quality:cheapThink |
CoreWeave | $0.125 per million tokens input · $1.11 per million tokens output |
$0.0137 per million tokens input · $0.126 per million tokens output
Credits pay the provider’s price plus 25%; cache hits cost less.
On a monthly plan, a typical call (8k tokens in, 1k out) uses about 0.2 plan credits, at the provider's own price.
cv11
Qwen models pair large context windows with solid multilingual ability, useful for roleplay that switches languages or references long lore documents.
With a 262,144-token context window, it can hold a long-running roleplay or a full lorebook in memory without losing earlier plot details, so it suits multi-session campaigns and character cards with heavy backstory.
It is priced low enough for high-volume daily roleplay without the per-message cost adding up quickly.
Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token. It uses a hybrid sparse mixture-of-experts architecture combining Gated...
Supplied by cv11, not written by Composite.
Qwen: Qwen3.8 Max Prime · Qwen: Qwen3.8 Omni Flash · Qwen: Qwen3.8 Max (0902)