Get started

Qwen3 30B A3B

Sold by 3 sellers, timed at 74.7 tok/s, on 5 Design Arena boards, with results on 1 openly licensed board, and not scored by Artificial Analysis.

retrieved sha256 · 0bb53a77713f

Models/Qwen/Qwen3 30B A3BUSD per 1M tokens · as OpenRouter routes it
qwen/qwen3-30b-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$0.12per 1M tokens
Output$0.50per 1M tokens
Cached inputnot listedper 1M tokens
Context41k40,960 tokens
What Vitna read · the provider's own pricing page, kept as a dated recordin / out per 1M tokens
DeepInfraopen weights$0.12 / $0.50confirmed 22 Nov 2026

The same weights, served by one provider at its own list price, read from deepinfra.com. OpenRouter's rate is its routed rate across whichever sellers it uses, and the two need not agree.

What the list says · the fields OpenRouter publishes for this id11 fields
AuthorQwen
Listed
Canonical idqwen/qwen3-30b-a3b-04-28
Context40,960 tokens
Max output16,384 tokens
Modalitiestext → text
TokenizerQwen3
Reasoningoptional, on by default
Moderationnot moderated by OpenRouter
Knowledge cutoff2025-03-31
What the weights say · read from the model's own repository on Hugging Faceretrieved from huggingface.co ↗
Parameters30.5B · 30,532,122,624 declared
Model classQwen3MoeForCausalLM
Downloadnot gated
Published

What sellers charge

3 sellers · 1.88x from cheapest to dearest

One tick per seller, on a log scale of input price per million from $0.096 at DeepInfra to $0.18 at DeepInfra. 1 seller undercuts the $0.12 listed; 1 is dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.

Measured speed

2,041 requests · 3 endpoints · 30-minute window
Throughput74.7 tok/sMedian output rate, weighted across the 3 endpoints serving it by the requests each answered.the speed board
First token724msMedian wait before the first token, over the same window and weighted the same way.how Vitna reads a figure
Endpoint spread64 to 76The 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

3 priced · 3 timed · cheapest first
  1. DeepInfradeepinfra/fp8/flex$0.096$0.40not listedfp837knot matchednonenonenot matchedno reading
  2. DeepInfradeepinfra/fp8 · listed$0.12$0.50not listedfp816knot matchednonenonenot matched100%
  3. DeepInfradeepinfra/fp8/priority$0.18$0.75not listedfp837knot matchednonenonenot matched99.36%
  4. DeepInfranot matchednot matchednot matchednot matchedno offer64 tok/s48853ms1 reqno offer
  5. DeepInfranot matchednot matchednot matchednot matchedno offer76 tok/s58784ms713 reqno offer
  6. DeepInfranot matchednot matchednot matchednot matchedno offer74 tok/s57691ms1,327 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: 2 of the 3 timed routes are slower than it, and the middle route runs at 74 tok/s. The widest median-to-p90 gap is DeepInfra's, 1.33x.

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 . 1 of the 3 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/priority) at 97.82% on 16 Nov 2026. 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

17 parameters · 3 sellers
  1. frequency_penaltyevery seller
  2. include_reasoningevery seller
  3. logit_biasevery seller
  4. max_tokensevery seller
  5. min_pevery seller
  6. presence_penaltyevery seller
  7. reasoningevery seller
  8. repetition_penaltyevery seller
  9. response_formatevery seller
  10. seedevery seller
  11. stopevery seller
  12. structured_outputsevery seller
  13. temperatureevery seller
  14. top_kevery seller
  15. top_pevery seller
  16. tool_choice1 of 3not DeepInfra (deepinfra/fp8/flex) and DeepInfra (deepinfra/fp8/priority)
  17. tools1 of 3not DeepInfra (deepinfra/fp8/flex) and DeepInfra (deepinfra/fp8/priority)

15 of these 17 are accepted by every one of the 3 sellers that publish a list; the other 2 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 $0.215 per million tokens, 3 in to 1 out: cheaper than 264 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

Each 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.

Intelligence Indexnot scoredArtificial Analysis has not scored this model on its Intelligence Index.the intelligence leaderboard
Coding Indexnot scoredArtificial Analysis has not scored this model on its Coding Index.the coding leaderboard
Agentic Indexnot scoredArtificial Analysis has not scored this model on its Agentic Index.the agentic leaderboard
  1. Code · Models930 Elowin rate 37.5% · source rank 140the board, 120 ranked
  2. Data visualisation · Models967 Elowin rate 39.0% · source rank 123the board, 110 ranked
  3. Game development · Models887 Elowin rate 33.9% · source rank 142the board, 120 ranked
  4. UI components · Models934 Elowin rate 42.4% · source rank 131the board, 116 ranked
  5. Websites · Models945 Elowin rate 37.7% · source rank 143the board, 126 ranked
Epoch Capabilities Index · a fit by a third party over every benchmark this model has results onthe whole index, with intervals
Index134.2131.3 to 136.7
Position147 of 258above 43% of the models placed
Results behind it6benchmark results
As Epoch names itQwen3-30B-A3Breleased 29 Apr 2025

Position 147 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 diamond342none0.50
unstated0.62
OTIS Mock AIME 2024-2025319none0.26
unstated0.63

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 Qwen3-30B-A3B_none and Qwen3-30B-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 30B 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.