Qwen3.6 35B A3B
Sold by 10 sellers, timed at 50.9 tok/s, scored 41.9 on the Coding Index, and with results on 1 openly licensed board.
retrieved sha256 · 0bb53a77713f
The same weights served by 3 providers, each at its own list price. OpenRouter's rate is its routed rate across whichever of them it uses, and it need not match any one of them.
What sellers charge
10 sellers · 4.8x from cheapest to dearestOne tick per seller, on a log scale of input price per million from $0.05 at Darkbloom to $0.24 at SiliconFlow. 2 sellers undercut the $0.10 listed; 3 are dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.
Measured speed
1,012,687 requests · 10 endpoints · 30-minute windowOpenRouter's own timing of its routed traffic, weighted by requests, over one 30-minute window read . Every route is on the endpoint ledger below.
Every endpoint
10 priced · 10 timed · cheapest first- Darkbloomdarkbloom/fp4$0.05$0.70$0.025fp4236k68 tok/s36906ms10,777 req99.86%
- DeepInfradeepinfra/fp8/flex$0.08$0.76$0.08fp866knot matchednonenonenot matched100%
- AkashMLakashml/fp8 · listed$0.10$0.90$0.05fp8236k23 tok/s191.4s53,305 req100%
- CoreWeavecoreweave/fp8 · listed$0.10$0.89$0.035fp8236k62 tok/s50499ms524,093 req99.93%
- DeepInfradeepinfra/fp8 · listed$0.10$0.95$0.10fp866knot matchednonenonenot matched99.71%
- Parasailparasail/fp8 · listed$0.10$0.90$0.025fp8236k54 tok/s37.1752ms232,176 req100%
- Venicevenice/fp8 · listed$0.10$1.00not listedfp866k22 tok/s18784ms129 req100%
- PrimeIntellectprimeintellect/fp8$0.11$1.01$0.099fp866k38 tok/s292.03s25,926 req100%
- DeepInfradeepinfra/fp8/priority$0.15$1.425$0.15fp866knot matchednonenonenot matched99.82%
- SiliconFlowsiliconflow/fp8$0.24$1.80$0.15fp8236k34 tok/s19978ms177 req94.09%
- DeepInfranot matchednot matchednot matchednot matchedno offer21 tok/s171.73s83,158 reqno offer
- DeepInfranot matchednot matchednot matchednot matchedno offer22 tok/s18774ms82,236 reqno offer
- DeepInfranot matchednot matchednot matchednot matchedno offer20 tok/s18573ms710 reqno offer
Each track runs from a route's p90, the rate nine in ten of its requests met or beat, up to its median. Weighted by requests, the figure above reports where the traffic went rather than what one route does: 7 of the 10 timed routes are slower than it, and the middle route runs at 28.5 tok/s. The widest median-to-p90 gap is Darkbloom's, 1.89x.
DeepInfra publishes 3 offers and had 3 routes timed. The two documents share no route id, so those offers and routes are printed apart rather than paired.
Uptime is OpenRouter's own figure for each seller over the trailing day, read once a day, and the column prints the latest reading, taken . 10 of the 13 sellers with a reading fell under 99% on at least one of the 23 days held, 23 Oct 2026 to 21 Nov 2026: AtlasCloud at 12.46% on 9 Nov 2026, SiliconFlow at 69.73% on 15 Nov 2026, DekaLLM at 78.66% on 11 Nov 2026, DeepInfra (deepinfra/fp8/flex) at 85.71% on 15 Nov 2026, and 6 more. Source: OpenRouter's public endpoints API, one reading a day for each seller: 1,211 sellers carried one on the latest of the 23 days held.
Source: OpenRouter's public endpoints API. 1257 sellers across 359 models when the mirror ran, 186 of them served by more than one. List prices, not what anybody paid; a discount is the seller's own. Read . Speed is OpenRouter's own timing over one 30-minute window, read .
What each seller accepts
19 parameters · 10 sellers- frequency_penaltyevery seller
- include_reasoningevery seller
- max_tokensevery seller
- presence_penaltyevery seller
- reasoningevery seller
- temperatureevery seller
- top_kevery seller
- top_pevery seller
- repetition_penalty9 of 10not Venice
- response_format9 of 10not Venice
- stop9 of 10not SiliconFlow
- structured_outputs9 of 10not Venice
- seed8 of 10not SiliconFlow and Venice
- tool_choice8 of 10not CoreWeave and SiliconFlow
- tools8 of 10not CoreWeave and SiliconFlow
- logit_bias5 of 10
- logprobs4 of 10
- min_p4 of 10
- top_logprobs4 of 10
8 of these 19 are accepted by every one of the 10 sellers that publish a list; the other 11 are declared by some sellers and not others, so what a request sending one of them gets depends on which seller serves it. Each list is the seller's own declaration in OpenRouter's endpoints API, read , and nothing here tested one.
Price position
342 priced modelsBlended $0.30 per million tokens, 3 in to 1 out: cheaper than 246 of the 342 priced models in the catalog. Each bar is one stretch of the catalog on a logarithmic price scale; the mark is this listing.
Measured quality
145 models with an index · 132 on a design arena boardEach mark is one board this model stands on, placed at the share of that board it scores above, and led to its number in the list under the rule. The columns and the ledger below carry the figures.
Not ranked in any Design Arena category.
Position 103 of the 258 models Epoch has placed, which is not the population of either board above, on a scale anchored at Claude 3.5 Sonnet 130 and GPT-5 150. Epoch states that absolute values are meaningless by themselves and that comparisons between models are what the index is for, that values are not linearly related to benchmark accuracy, and that a highly specialised model may score low while being capable in its own domain.
Benchmark results
1 of 6 boards · 7 results| Task | All runs | Effort | Score |
|---|---|---|---|
| FrontierMath-Tiers-1-3-v2-Private | 128 | none | 0.20 |
| unstated | 0.18 | ||
| GPQA diamond | 342 | none | 0.85 |
| unstated | 0.84 | ||
| Mystery Game Puzzles | 149 | none | 0.22 |
| OTIS Mock AIME 2024-2025 | 319 | none | 0.69 |
| unstated | 0.87 |
Mean accuracy, 0 to 1, on the task the row names, as Epoch AI runs it and averaged over repeated runs. All eighteen of their tasks, including the held-out FrontierMath tiers, named as they name them. The model id carries the effort setting the run used. Epoch AI Benchmarking Hub names these runs qwen3.6-35b-a3b_none and qwen3.6-35b-a3b, and the effort beside each score says which run it is, unstated where the publisher wrote none. A run missing from a task's rows has no published result there, which is not a score of zero. All runs is how many runs the publisher scored on the task, every model's. Read from epoch.ai ↗, which dates its own data . Epoch AI Benchmarking Hub data, by Epoch AI, under CC BY 4.0.
1 of the 6 openly licensed boards Vitna mirrors carries a result for Qwen3.6 35B A3B, each printed as its publisher published it and none added to another. MTEB, LiveBench, HELM, Vectara hallucination leaderboard and Aider polyglot have not published one under a name that matches its listing, and names are matched to the catalog exactly and never guessed. A task that is a link opens its column on /capability, with every run scored on it; an effort that is a link opens its run's row there, where the run is ranked on Vitna's scale. Every board, with its licence and what its publisher measures, is on /benchmarks.
Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Source: each model's own Hugging Face repository (huggingface.co), as of 20 Nov 2026. 162 repositories read, 4 did not answer.
Licence, parameter count, model class and gating: each model's own Hugging Face repository, the one OpenRouter names for the listing.
A licence is the tag the uploader put on the repository, read from nothing else and verified by nobody.
A parameter count is the total the repository's safetensors files declare; for a mixture of experts that is every expert, not the share one token runs through.
Source: the Epoch AI Capabilities Index (epoch.ai/eci), as of 20 Nov 2026. 258 models placed over 2917 benchmark results.
Epoch AI Benchmarking Hub data, by Epoch AI, under CC BY 4.0.
Vitna ran none of these evaluations and adjusted none of these figures.
