Free sweep, 2026-11-03: what changed, and what could add capacity #369
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
The sweep of free plans, 2026-11-03. Some numbers come from community lists, marked below. New models need a recall probe before they join a chain.
The problem is capacity. Run 170 asked for about 410 strong requests an hour, averaging 54K input tokens (469 requests and 25.2M input tokens in 64.5 minutes, before most of its steps ran).
beam-freeSo the steady supply is about 220 an hour against 410 wanted, and a second run the same day gets less.
Changed since our catalog (2026-10-17)
longcat-2.5-preview-freeis gone. New:beam-free,mimo-v2.6-flash-free,flint-40b-free,pulse-alpha-free, two Nemotrons. Unknown whether the block is per model.Qwen3.8-27BandQwen3.6-27B(262K). Worth testing whether its 2 a minute is per model.mercury-3(1M). Our probe failed on 10-17 at capacity, so it's off. Key already on production.glm-4.7-flashalso failed its probe as overloaded, not on recall. Retest.dots-studio/dots-3-note-preview:free; ours lacks the prefix, so check it.inklingfree models end 11-04.gemini-3.6-flashis new on the free tier.Worth adding
Qwen3.8-27B(262K, images)Hetzner alone (600 an hour) covers the demand, and it's the model production's
mainprofile uses, so a run could stay on one model.Later, or unknown: Vercel AI Gateway (three $0 models, limits unpublished), Amazon Nova (free tier, numbers not found), Requesty (200 a day).
Not worth it: AIHubMix (100 a day, shared), IteraCompute (131K window, 8K output), Dahl (one-time, data use unstated, unknown GPU hosts), ModelScope (real-name check), Cerebras (card), and the many lists now marked no longer free (GitHub, Together, SambaNova, Hugging Face, DeepInfra, DeepSeek, Chutes).
Avoid: LLM7 and similar routers offering paid frontier models free, with little or no sign-up. Who pays, and who reads the prompts, can't be checked.
Done when
Sources: Hetzner, Kilo, Zen, OpenRouter, Gemini, Cloudflare, Cohere, Groq, Alibaba, Inception, Vercel, models.dev, community: free-llm.com.
Corrections after checking the sign-up pages
experiments.hetzner.com/inference.:freemodel with one. So it may or may not work; test it.keyenv.pyonly knowsKILO_API_KEYso far.HETZNER_API_KEY,MISTRAL_API_KEYandNOUS_API_KEYneed adding there, and all of them indeploy/docker-compose.yml, before production can use them.Hetzner's token works (2026-11-03). One tiny request to
Qwen3.8-27Breturned 200, and it answered asQwen/Qwen3.8-27B. Its list showsQwen/Qwen3.6-35B-A3B,Qwen/Qwen3.6-35B-A3B-FP8,Qwen/Qwen3.8-27BandQwen3.8-27B, so the ids differ from the docs: check which to use.The account needed a payment method (PayPal was accepted, and its payments are manual) and a government ID. Limit headers: none returned, so the 10 a minute is still only what the docs say. Recall and tool probes are still to do.
Mistral is out (2026-11-03). On our account, the API keys page still says "Your organization is on the Free plan. Upgrade to activate your API keys and access the full API", with New key disabled. Mistral's quickstart says the opposite, and looks out of date: its API keys page says you cannot create a Free mode key. Two others report the same since October (one, two). Community lists still show about 1B tokens a month, which seems to reflect only older accounts with existing keys.
It's paid only now, and not worth it: a run like run 170 would cost about $13 in input tokens alone at Mistral Large's listed price.
So Hetzner is the only large new source, and it's experimental. That makes a second one, such as Nous's free Beam, worth more.
Nous is out, and Kilo's key works (2026-11-03).
beam-free. Whether it lifts the 100 an hour is still untested, and its docs say the limit stays per IP (#367).Where that leaves new capacity: Hetzner is the only large new source. Its backups are the no-account retests: OVH's other Qwen models, Inception and Z.ai. Vercel and Amazon Nova stay in reserve if those fall short.
Nous confirmed out (2026-11-03). The $0 checkout works and puts the account on the Free plan, but API keys still can't be created. The plan page explains it: "API keys always use your credits (pay-as-you-go), not included usage", and the Free plan has $0 monthly credits. The plan is for Nous's own Hermes Agent app, signed in through OAuth. Pointing Braid at that sign-in would be using one product's access for another, so it's not an option.
Gemini's real free limits, read from AI Studio (2026-11-03). They match the catalog's third-party numbers, so those can be marked as read from the source.
gemini-3.5-flash-litegemini-3.1-flash-litegemini-3.8-flashgemini-3.6-flashgemini-3.1-flash-liteandgemini-3.8-flashhave no token limit recorded; add 250K.gemini-3.6-flash: only another 20 a day, so it could help the judge tier at most.Retests with tiny requests (2026-11-03)
Qwen3.8-27BandQwen3.6-27Badd capacity with no key. It can be slow: one request timed out at 120 s, one took 25 s.mercury-3andmercury-2.5answered in 0.5 s. The 10-17 probe failure is out of date, so probe again.glm-4.7-flashstill answers "overloaded" (1305). It stays out.deepseek-v4.1-flash-free,beam-freeandmimo-v2.6-flash-freeall answeredFreeUsageLimitError. The block covers every model, so its new free models add no capacity.dots-3-note-preview:freestill answers, as an alias of the prefixed id. No fix needed.The plan is in the OpenSpec change
free-capacity-hetzner.Measured: Hetzner is steady but slow (2026-11-03). I replayed run 170's spec-test step with Hetzner at the head of the chain, and stopped it at 31 minutes with no file written yet.
Hetzner refused nothing. Two of its requests thought for about 8,000 tokens, 6.4 and 7.5 minutes each.
Its 10 a minute isn't the limit that counts: at 86 s a request, one session gets about 40 requests an hour.
So the order changes. Kilo leads the strong and reader chains while its count lasts, and Hetzner carries a step when Kilo refuses, ahead of OVH. A private repository can't use Kilo, so Hetzner leads there.
Shipped in PR 371. Left open until a full Free run shows it holds, which is the plan's next step.
In the chains now:
In the catalog: Gemini's limits as AI Studio shows them, Zen's block at about 180 requests, and Zen's longcat-2.5-preview-free ended.
Same step, same capsule, two chain orders:
Left out, or not worth it: Mistral and Nous need payment for API keys, whatever their pages suggest. Gemma on Gemini's API has too few tokens a minute. Z.ai is still overloaded.
Owner decisions, not urgent: OpenRouter's one-time $10 for 1,000 requests a day, and NVIDIA's opt-in, whose terms don't allow auto-approve to stay on.
Closing: run 171 went end to end on the new chains, through the Visual check to a merged PR, in 104.7 minutes with nobody needed. Run 170 had stopped at its first spec-test step after 64.5.
Kilo didn't reach its count on this run, so running out rests on the sim
free-kilo-refuses-early.Left with the owner, not urgent: OpenRouter's $10 one-time for 1,000 a day, and NVIDIA's opt-in. Details of the run are on #365.