Get started

Qwen3.5-35B-A3B

Sold by 7 sellers, timed at 62.6 tok/s, scored 37.0 on the Coding Index, and with results on 1 openly licensed board.

retrieved sha256 · 0bb53a77713f

Models/Qwen/Qwen3.5-35B-A3BUSD per 1M tokens · as OpenRouter routes it
qwen/qwen3.5-35b-a3bQwen · base listing · Qwen3 tokenizer
OpenRouter list price · the routed rate, as the catalog carried itretrieved 17 Nov 16:29 UTC from openrouter.ai ↗
Input$0.14per 1M tokens
Output$1.00per 1M tokens
Cached input$0.05per 1M tokens
Context262k262,144 tokens
What Vitna read · 3 providers, each from its own pricing page, kept as dated recordsin / out per 1M tokens
DeepInfraopen weights$0.14 / $1.00confirmed 22 Nov 2026
Parasailopen weights$0.15 / $1.00confirmed 22 Nov 2026
Novita AIopen weights$0.25 / $2.00confirmed 22 Nov 2026

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 the list says · the fields OpenRouter publishes for this id10 fields
AuthorQwen
Listed
Canonical idqwen/qwen3.5-35b-a3b-20260224
Context262,144 tokens
Max output81,920 tokens
Modalitiestext, image, video → text
TokenizerQwen3
Reasoningoptional
Moderationnot moderated by OpenRouter
What the weights say · read from the model's own repository on Hugging Faceretrieved from huggingface.co ↗
Parameters36B · 35,951,822,704 declared
Model classQwen3_5MoeForConditionalGeneration
Downloadnot gated
Published

What sellers charge

7 sellers · 3.69x from cheapest to dearest

One tick per seller, on a log scale of input price per million from $0.065 at Darkbloom to $0.24 at SiliconFlow. 2 sellers undercut the $0.14 listed; 4 are dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.

The same sellers grouped by the number format each declares, on the same scale. The fp8 group alone spans 58% of the range, so the cheap end is not cheap for its number format. A declaration is checked by nobody: this groups the rows and certifies none.

Measured speed

22,240 requests · 7 endpoints · 30-minute window
Throughput62.6 tok/sMedian output rate, weighted across the 7 endpoints serving it by the requests each answered.the speed board
First token496msMedian wait before the first token, over the same window and weighted the same way.how Vitna reads a figure
Endpoint spread22 to 128The slowest and the fastest route serving this model, in tokens per second. Speed is a property of the route, not of the weights.the board, fastest first

OpenRouter'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

7 priced · 7 timed · cheapest first
  1. Darkbloomdarkbloom/fp4$0.065$0.75$0.0325fp4131k72 tok/s53507ms3,139 req99.73%
  2. DeepInfradeepinfra/fp8/flex$0.112$0.80$0.04fp882knot matchednonenonenot matched91.66%
  3. DeepInfradeepinfra/fp8 · listed$0.14$1.00$0.05fp882knot matchednonenonenot matched99.06%
  4. Parasailparasail/fp8$0.15$1.00$0.05fp8236k122 tok/s85.5434ms962 req99.31%
  5. Venicevenice$0.15$1.00$0.05not declared16k88 tok/s55750ms1,154 req99.85%
  6. Alibabaalibaba$0.1625$1.30not listednot declared66k128 tok/s63560ms1,464 req100%
  7. SiliconFlowsiliconflow/fp8$0.24$1.80$0.15fp8236k46 tok/s31.2593ms265 req94.37%
  8. DeepInfranot matchednot matchednot matchednot matchedno offer49 tok/s28471ms15,255 reqno offer
  9. DeepInfranot matchednot matchednot matchednot matchedno offer22 tok/s11549ms1 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: 3 of the 7 timed routes are slower than it, and the middle route runs at 72 tok/s. The widest median-to-p90 gap is Alibaba's, 2.03x.

DeepInfra publishes 2 offers and had 2 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 . 6 of the 8 sellers with a reading fell under 99% on at least one of the 23 days held, 23 Oct 2026 to 21 Nov 2026: DeepInfra (deepinfra/fp8/flex) at 0.00% on 19 Nov 2026, SiliconFlow at 48.14% on 25 Oct 2026, AtlasCloud at 62.22% on 20 Nov 2026, Darkbloom at 66.84% on 6 Nov 2026, and 2 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 · 7 sellers
  1. frequency_penaltyevery seller
  2. include_reasoningevery seller
  3. max_tokensevery seller
  4. presence_penaltyevery seller
  5. reasoningevery seller
  6. response_formatevery seller
  7. structured_outputsevery seller
  8. temperatureevery seller
  9. top_kevery seller
  10. top_pevery seller
  11. stop6 of 7not SiliconFlow
  12. repetition_penalty5 of 7not Alibaba and Venice
  13. seed5 of 7not SiliconFlow and Venice
  14. tool_choice5 of 7not DeepInfra (deepinfra/fp8) and SiliconFlow
  15. tools5 of 7not DeepInfra (deepinfra/fp8) and SiliconFlow
  16. logprobs4 of 7not DeepInfra (deepinfra/fp8), DeepInfra (deepinfra/fp8/flex), and SiliconFlow
  17. top_logprobs4 of 7not DeepInfra (deepinfra/fp8), DeepInfra (deepinfra/fp8/flex), and SiliconFlow
  18. logit_bias3 of 7only DeepInfra (deepinfra/fp8), DeepInfra (deepinfra/fp8/flex), and Parasail
  19. min_p2 of 7only DeepInfra (deepinfra/fp8) and DeepInfra (deepinfra/fp8/flex)

10 of these 19 are accepted by every one of the 7 sellers that publish a list; the other 9 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 models

Blended $0.355 per million tokens, 3 in to 1 out: cheaper than 237 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 board
Intelligence Indexnot scoredArtificial Analysis has not scored this model on its Intelligence Index.the intelligence leaderboard
Coding Index37.0Position 78 of 118, above 34% of the models scored.the coding leaderboard
Agentic Indexnot scoredArtificial Analysis has not scored this model on its Agentic Index.the agentic leaderboard

Not ranked in any Design Arena category.

Epoch Capabilities Index · a fit by a third party over every benchmark this model has results onthe whole index, with intervals
Index144.1141.5 to 146.6
Position104 of 258above 60% of the models placed
Results behind it4benchmark results
As Epoch names itQwen3.5-35B-A3Breleased 24 Feb 2026

Position 104 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 · 4 results
Epoch AI Benchmarking Hub · publisher-run, CC-BY-4.02 of 22 tasks
TaskAll runsEffortScore
GPQA diamond342none0.81
unstated0.83
OTIS Mock AIME 2024-2025319none0.70
unstated0.54

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 qwen3.5-35b-a3b_none and qwen3.5-35b-a3b, and the effort beside each score says which run it is, unstated where the publisher wrote none. 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 Qwen3.5-35B-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. Every board, with its licence and what its publisher measures, is on /benchmarks.

Sources

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.