Qwen3 235B A22B Thinking 2507 API pricing
Alibaba · Mid-range price tier · Apache 2.0 · released 2025-07. Prices in USD per 1 million tokens. Data snapshot: 2026-11-12.
A support ticket workload of 100K requests a month costs $242.50 on Qwen3 235B A22B Thinking 2507; the next cheaper model in the Mid-range price tier for this workload, Glimpse, does it for $240.00. The price we quote for Qwen3 235B A22B Thinking 2507 has changed once since 2026-08-15, most recently on 2026-10-10 when the quoted host changed (+54.0%); it is now the 40th cheapest of 113 models in the Mid-range price tier. Only one host, Venice, sells it, so there is no second price to shop. It is rated a stable release but does not meet all four criteria of the FinOps Friendly badge.
- Input: $0.45 per 1M tokens
- Output: $3.50 per 1M tokens
- Batch API (in/out): Not published
- Prompt-cache read: Not published
- Context window: 128K tokens
- Arena ELO: 1415 as of 2026-11-08
Market position
On blended price ($2.585 per 1M tokens at 30% input, 70% output) it is cheaper than 65% of models in the Mid-range price tier: 40th cheapest of 113. Across every price tier it is cheaper than 34% of priced text models in the catalogue: 200th cheapest of 303. Its Arena ELO of 1415 ranks 86th of the 160 models that carry one, above 47% of them. With an Arena score of 1415, Qwen3 235B A22B Thinking 2507 sits below OptimNow's lowest working quality threshold, Standard (Arena 1420+), so the quality-bar method would not shortlist it for any of the 8 use cases the table prices. It is 5 points under the Standard bar (Arena 1420+), within the benchmark's 20-point noise. Each threshold is a distance below the best ranked model, which scores 1530 on Arena scores as of 8 Nov 2026.
Licence, size and capabilities
Qwen3 235B A22B Thinking 2507 is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token. Its published size is 235B parameters. Listed capabilities: Text, Reasoning, Agents. Filling its 128K context window once costs $0.0576 in input tokens at list price.
Pricing by host
A host is a company that runs the model and sells it by the token: the model maker, a cloud platform like Amazon Bedrock, Azure or Vertex, or an independent host. One model, several hosts, and often several prices. This model is sold by 1 host. The price we quote elsewhere on this page is marked. We quote one host's rate so that models stay comparable; it is not a recommendation to buy there.
Venice is the only host with a published rate, $0.45 input / $3.50 output per 1M tokens, which is the price we quote. It serves a quantized fp8 build, which is not the same product as the full-precision model.
| Host | Input $/1M | Output $/1M | Precision |
|---|---|---|---|
| Venice (price we quote) | $0.45 | $3.50 | fp8 |
List prices as published on 2026-11-12, via OpenRouter. Negotiated rates, committed-use discounts and provisioned capacity are not shown.
Monthly cost at 100K requests, by business use case
From the after-discounts figure (prompt caching on the cached input share, batch API for async workloads, only where Alibaba publishes those rates), which assumes you implement both, to the list price. One figure means no discount is published.
| Use case | Tokens (in/out) | Monthly (after discounts – list) |
|---|---|---|
| Support Ticket | 1,500 / 500 | $243 |
| Knowledge Q&A | 2,000 / 800 | $370 |
| Meeting Summary | 10,000 / 1,200 | $870 |
| Marketing Content | 2,500 / 1,800 | $743 |
| Coding Task | 3,000 / 2,000 | $835 |
| Invoice Processing | 1,500 / 600 | $277 |
| Call Summary | 2,000 / 700 | $335 |
| Agent Workflow | 6,000 / 3,000 | $1.3K |
Price history
Qwen3 235B A22B Thinking 2507 has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.23 input / $2.30 output per 1M tokens. The price we quote has changed once since then, and that was a change of quoted host rather than a repricing. The latest change, on 2026-10-10, was +54.0% on the blended price (30% input, 70% output), to $0.45 input / $3.50 output (quoted host changed, not a repricing).
| Date | Old price (in / out) | New price (in / out) | Blended change |
|---|---|---|---|
| 2026-10-10 | $0.23 / $2.30 | $0.45 / $3.50 | +54.0% (quoted host changed, not a repricing) |
Alternatives to Qwen3 235B A22B Thinking 2507
2 of the 4 comparable models below cost less than Qwen3 235B A22B Thinking 2507: Qwen3.5-122B-A10B undercuts it by 41%, and scores 2 Arena points higher.
- Next cheaper model in the Mid-range price tier: Kimi K2.6 (Moonshot) at $2.582 blended, under 1% less.
- Next more expensive: Kimi K2.7 Code (Moonshot) at $2.63 blended, 2% more.
The nearest models in the Mid-range price tier on Arena ELO, each with its blended price and score difference against Qwen3 235B A22B Thinking 2507.
- Qwen3.5-122B-A10B (Alibaba): $0.26 in / $2.08 out per 1M, Arena ELO 1417 (41% cheaper, 2 ELO points higher)
- Gemini 2.5 Flash (Google): $0.30 in / $2.50 out per 1M, Arena ELO 1417 (29% cheaper, 2 ELO points higher)
- GPT-5.4 Mini (OpenAI): $0.75 in / $4.50 out per 1M, Arena ELO 1412 (31% more expensive, 3 ELO points lower)
- GPT-5.2 (OpenAI): $1.75 in / $14.00 out per 1M, Arena ELO 1412 (299% more expensive, 3 ELO points lower)
- Cheapest within 15 Arena points: Step 3.5 Flash (StepFun), 91% cheaper, 12 ELO points lower, out of 31 models in that band.
- Highest Arena ELO within 20% of this price: Gemini 3.8 Flash (Google), 10% more expensive, 85 ELO points higher, out of 13 models in that price band.
Other models from Alibaba
The 8 nearest to Qwen3 235B A22B Thinking 2507 in price, of 39 other Alibaba models in the catalogue. Prices per 1M tokens, input then output.
- Qwen3.6 Plus: $0.325 / $1.95
- Qwen3.5-122B-A10B: $0.26 / $2.08
- Qwen3.5-27B: $0.26 / $2.60
- Qwen3.6 27B: $0.32 / $3.20
- Qwen3.5 397B A17B: $0.60 / $3.60
- Qwen3.7 Max: $1.475 / $4.425
- Qwen3.8 2.4T A95B: $2.00 / $6.00
- Qwen3.8 Max (0902): $2.00 / $6.00
All 40 Alibaba models and prices
Common questions about Qwen3 235B A22B Thinking 2507 pricing
- How much does Qwen3 235B A22B Thinking 2507 cost?
- Qwen3 235B A22B Thinking 2507 costs $0.45 per 1M input tokens and $3.50 per 1M output tokens at Venice, the host whose rate we quote. Venice serves a quantized fp8 build, which is not the same product as the full-precision model. Prices as listed on 2026-11-12.
- Which host is cheapest for Qwen3 235B A22B Thinking 2507?
- Venice is the only host with a published rate, $0.45 input / $3.50 output per 1M tokens, which is the price we quote. It serves a quantized fp8 build, which is not the same product as the full-precision model. Host rates as listed on 2026-11-12.
- Does Qwen3 235B A22B Thinking 2507 offer batch or cached pricing?
- No. Neither a batch rate nor a prompt-cache rate is published for Qwen3 235B A22B Thinking 2507 as of 2026-11-12.
- What would Qwen3 235B A22B Thinking 2507 cost per month?
- With the Support Ticket profile (1,500 input and 500 output tokens per request), 100,000 requests a month cost about $243 on Qwen3 235B A22B Thinking 2507 at the list price quoted on 2026-11-12.
- Can Qwen3 235B A22B Thinking 2507 be self-hosted?
- Yes. Qwen3 235B A22B Thinking 2507 is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token.
- Has the price of Qwen3 235B A22B Thinking 2507 changed?
- Qwen3 235B A22B Thinking 2507 has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.23 input / $2.30 output per 1M tokens. The price we quote has changed once since then, and that was a change of quoted host rather than a repricing. The latest change, on 2026-10-10, was +54.0% on the blended price (30% input, 70% output), to $0.45 input / $3.50 output (quoted host changed, not a repricing).
Its weights are published, which changes what you may run yourself: the cost gap between open weights and proprietary APIs.
These are list prices. Your real cost also depends on caching and batching, negotiated discounts, marketplace agreements, and the contracts you already have. OptimNow audits exactly that.
Build the business case · Get a FinOps review
LLM pricing · Models by maker · Cloud compute pricing · Compare models · Price barometer · FinOps guides · API and MCP server