Release watch: Qwen3.8

What exists, verified

Each row carries exactly one status label. Shipped and verified means we saw the actual weight files at a named revision with a license. API only means the model answers requests but no weights exist. Announced means a vendor promise with no artifact. Not official means artifacts exist under the name but none come from the maker. A repository name is never proof of a release.

One unverified report claims the coming weights will carry geographic license restrictions. It is a single source and corroborated nowhere; we treat it as rumor until the license file exists, and note it only so its later appearance or absence is on the record.

The name-squat hazard, before the weights exist

Within days of the announcement, and before any real weights, repositories using the Qwen3.8 name appeared on public model hubs under third-party accounts. The ones we examined fall into two shapes: placeholder cards that describe planned quantizations but contain no weight tensors at all, and rebadges, where a small model distilled from or unrelated to the announced ones circulates under the 3.8 name. Secondary articles have already cited some of these as if they were releases. The two are different hazards with the same conclusion: an empty card cannot be a release, and real tensor files under the name are still someone else's model, not the announced Max or 27B.

When the drop happens, the safe path is boring: take the repository link from the Qwen team's own announcement channels, or the official organization page directly, confirm the license file is present, confirm the weight shards are actually there with sizes that make sense for the stated parameter count, and verify checksums after downloading. We will link the verified repositories from this page the day they exist, with revisions and sizes as we confirm them.

First measurements of the API model

On 2026-08-09 we ran a fixed fifteen-call battery against Qwen3.8-Max through one aggregator route (OpenRouter), with every raw request and response body retained. This is one serving route on one afternoon, one to three calls per cell: diagnostics, not a benchmark. Total cost of the whole battery: seven cents, at the listed two dollars per million input tokens and six per million output.

The quick facts first. Asked what it is, the model answered "Qwen, also known as Tongyi Qianwen," built by Alibaba's Tongyi Lab. It solved the classic bat-and-ball trap correctly, produced a correct interval-merging function with three test assertions including the empty-list edge case, returned schema-conforming JSON on all three attempts, and passed our honesty probe cleanly: asked to summarize a file we never provided and report a score on a benchmark that does not exist, it said it had no such file and could not state the score, and asked for the data instead of inventing either.

The finding that will matter to anyone building on it: the model thinks by default, its thinking bills as output tokens inside your token cap, and it will happily spend your entire cap thinking and hand back an empty answer. This is a family trait we know well from serving two Qwen3.5 models locally (their field card is linked below), and it is fully present on this frontier API route:

Token caps are the request's max-tokens value on this route, which covers thinking plus the visible answer. "Thinking spent" is the reasoning-token count the API reported back. Every empty row finished with the length stop reason: the cap ran out before the first visible word.

Read the middle rows carefully: a medium reasoning puzzle exhausted a 2,000-token cap without producing one visible word, and a simple story request burned 1,200 tokens of thinking three times in a row and never wrote the story. The practical rule we use locally holds here: give thinking Qwens thousands of tokens of headroom, or use the effort control, and treat an empty answer with a length stop as a budget failure, not a model refusal. You pay for the thinking either way: our three empty story attempts cost more than the completed effort-controlled runs that actually answered.

End-to-end latency on this route, for scale: about three to five seconds for a short factual answer, about twelve for the JSON asks, about twenty-five to forty-six seconds when thinking runs long. Wall-clock through an aggregator, labeled as such.

What we will do the day the weights land

This machine already serves two Qwen3.5 mixture-of-experts models on one 32 GB graphics card, so the absorption path is rehearsed, and the runbook is written before the release rather than after. In order:

• Verify before downloading. Official repository, license file read, architecture string from the config, shard list with sizes. If the architecture is new to the runtimes, we say plainly that it does not run locally yet and track the support work instead of forcing it.
• Feasibility before the big one. The 27B-class model is a routine download. A Max-class open release is not: we compute storage, memory and rollback from the actual file listing first. We have run a model far larger than system memory before, streaming from disk, so "too big to be sensible" is a measurement question here, not a reflex. Loadable, able to generate, worth studying, and worth recommending are four different findings and we will label which one we reached.
• Checksums, then a first safe load at conservative settings, nothing else running.
• The same fixed battery we run on every model: identity and file accounting, load and memory placement, warm speed medians, the thinking-budget map from section 03 run locally with identical prompts, the reasoning toggle, tool calls, structured output, an honesty probe, and a long-context retrieval check. Same prompts as the API run above, so local-versus-hosted lands on day one with nothing invented.
• This page updates the same day, dated, with whatever we actually measured and nothing more.

Where this page comes from

Graphometer runs the current Qwen generation for real: the Qwen3.5 122B and 397B serve on one desktop here, measured in a field card whose numbers trace to dated recorded runs, including a speculative-decoding gate trap that the guides we used while serving those models do not call out. That working familiarity is the lens for this page: we are not previewing benchmarks, we are preparing to operate the new models and to publish what operating them is actually like.

Updates

Every existence claim on this page is dated. If today is later than the newest row above, treat the status table as unverified since that date and check the official organization directly.

Sources

• The Qwen team's 2026-08-03 announcement and the official Qwen organization on Hugging Face, checked 2026-08-09: no Qwen3.8 repositories under the official organization as of that date. ModelScope, the other named release venue, could not be enumerated from here, and no official 3.8 listing there surfaced through secondary sources either; we will check it directly when the drop lands.
• OpenRouter's public model API, 2026-08-09: Qwen3.8-Max listed at a one-million-token context, two dollars per million input and six per million output.
• Our fifteen-call recorded battery against Qwen3.8-Max via OpenRouter, 2026-08-09: raw request and response bodies, a per-call summary table, and spend accounting are project records that can be produced if questioned.
• Runtime repositories (llama.cpp, vLLM, SGLang, Ollama), checked 2026-08-09: Qwen3.5 support is merged and shipping; no merged Qwen3.8 support in those repositories as of that date.
• The Qwen3.5 pair field card on this site for the local serving record this page builds on.

Vendor claims on this page are labeled as claims. Model names and trademarks belong to their owners. This page contains no scripts, no tracking and no cookies, like the rest of this site.