# Qwen3.6 35B-A3B is now 1.5x faster on CARouter Cloud > Our coding model shipped with a built-in helper that drafts the next tokens for the main model to check. We switched it on with one line of configuration: 115 tokens a second for one user instead of 75, and 479 instead of 342 with sixteen requests at once, on the same DGX Spark. - Published: 2026-10-02 - Author: Flavien Berwick - Category: Product - Web page: https://carouter.ai/blog/qwen3-6-mtp-head-one-and-a-half-times-faster - In French: https://carouter.ai/fr/blog/qwen3-6-mtp-head-one-and-a-half-times-faster.md ![Now faster: Qwen3.6 35B-A3B is now 1.5x faster on CARouter Cloud. One request: 115 tokens a second, 75 before. Sixteen at once: 479, 342 before. A long coding answer: 41 seconds, 63 before. Cost: 2.3 GiB of memory.](https://carouter.ai/blog/qwen3-6-mtp-head-one-and-a-half-times-faster/illustration.png) Five weeks ago we put a faster model behind CARouter Cloud's coding endpoint. It came with a built-in speed feature that we had not switched on. Now it is on: the same model, on the same machine, at the same price, writes about one and a half times faster. The change was one line of configuration. [The previous article: why we now serve Qwen3.6 35B-A3B instead of Qwen3.8 27B.](https://carouter.ai/blog/qwen3-6-35b-a3b-instead-of-qwen3-8-27b) ## The short version - **What a client sees.** 115 tokens a second instead of 75. A long coding answer arrives in 41 seconds instead of 63. - **What we changed.** One line. It switches on a small helper already inside the model file, which drafts the next three tokens for the main model to check. - **When the machine is busy.** With sixteen requests at once, 479 tokens a second in total instead of 342. - **What it costs.** 2.3 GiB of memory out of 121. No new hardware and no new download. - **How we tested it.** On a rented identical machine, for 1.03 US dollars. Production restarted once, for six minutes. ## A helper that drafts, a model that checks A language model writes one token at a time, a token being a word or a piece of one. Each token costs one full pass of the model. With speculative decoding, a small helper drafts the next few tokens and the large model checks them all in one pass. It keeps the ones it agrees with and rewrites the first one it does not. The large model has the last word on every token, so the helper saves time without lowering quality. Qwen3.6 35B-A3B ships with such a helper inside its own file, called a multi-token-prediction head. We had missed it when we evaluated the model in August. ## What we measured Same machine, same model file, same memory budget. We measured, restarted the endpoint once with the helper on, and measured again the same evening. ![Tokens written per second on one DGX Spark, with and without the helper: 115 against 75 for one request, 479 against 342 for sixteen.](https://carouter.ai/blog/qwen3-6-mtp-head-one-and-a-half-times-faster/01-decode-by-streams.png) The helper drafts three tokens at a time and about seven in ten are accepted. Each pass yields about three tokens instead of one and costs about twice as much, which makes one and a half times faster. The gain is 1.53 times for one user and 1.40 times for sixteen. Across everything we measured it stays between 1.35 and 1.58 times. ## What that means for someone waiting ![Seconds until the answer is complete, without and with the helper: 63 and 41 for a long coding answer.](https://carouter.ai/blog/qwen3-6-mtp-head-one-and-a-half-times-faster/02-seconds-to-answer.png) A short reply arrives two seconds sooner and a long coding answer twenty-two seconds sooner. With eight people on the endpoint, each gets that long answer in 99 seconds instead of 145. A coding agent calls the model hundreds of times for one task. A third less waiting on every call adds up. ## What it costs The endpoint holds 2.3 GiB more memory, which the machine had to spare. Reading the prompt is 1% slower. It also remembers a little less conversation at once: 4.8 conversations of the maximum length instead of 5.5. Our traffic is far below that, so we changed no limit. ## How we tested it without switching anything off Every restart of the coding endpoint is six minutes during which it answers nobody. Trying each setting in production would have meant four of them. So we rented a machine with the same chip, by the hour, and ran our own software on it unchanged. We tried the helper with one, two and three tokens. Three won, and the evening cost 1.03 US dollars. ![How much faster the endpoint is with the helper, on the rented machine and on ours: between 1.40 and 1.62 times on both.](https://carouter.ai/blog/qwen3-6-mtp-head-one-and-a-half-times-faster/03-rehearsal-predicted-production.png) Then we changed one line in production and restarted once. The rented machine had predicted the result to within a few percent. ## What we cannot tell you We did not run the coding benchmark again. The large model still decides every token, and we checked that conversation, reasoning, tool calls and images all still work. We cannot show two identical answers. With all randomness switched off, this software does not always repeat itself, with the helper or without it. We found no difference that the helper causes, and we cannot prove there is none. One faster option remains, a computation routine written for this chip. It has open defects in the software we depend on, so we are not running it. ## What is running now Qwen3.6 35B-A3B with its prediction helper, on the same GB10, under the same name, qwen3.6-35b-a3b, at the same price. Context, images, tools and reasoning are unchanged. --- More articles: [CARouter Blog](https://carouter.ai/blog.md). What CARouter is and how to call it: [llms.txt](https://carouter.ai/llms.txt).