Qwen3 235B A22B Instruct 2507 model profile
Evidence snapshot:
Qwen3 235B A22B Instruct 2507 is a Alibaba Qwen model profile with a 262,144-token context window. The dated catalog snapshot records text modalities and interfaces for structured output, text generation, tool calling.
Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following, logical reasoning, math, code, and tool usage. The model supports a native 262K context length and does not implement "thinking mode" (<think> blocks). Compared to its base variant, this version delivers significant gains in knowledge coverage, long-context reasoning, coding benchmarks, and alignment with open-ended tasks. It is particularly strong on multilingual understanding, math reasoning (e.g., AIME, HMMT), and alignment evaluations like Arena-Hard and WritingBench.
Qwen3 235B A22B Instruct 2507 is a reference profile: Alibaba Qwen publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call.
Qwen3 235B A22B Instruct 2507 is an independent knowledge profile. It does not claim that Kendr hosts the model, quote a Kendr customer price, promise API availability, or advertise a Kendr routing receipt.
Qwen3 235B A22B Instruct 2507 accepts text and returns text. The snapshot records structured output, text generation, and tool calling as capabilities or interfaces.
The dated reference snapshot lists $0.0875 input and $0.35 output per million tokens. Cached input is $0.0175 per million tokens.
The 2026-09-06 snapshot includes 6 Design Arena categories and AutoExacto results for 13 provider observations. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable.
Qwen3 235B A22B Instruct 2507 ranked #83 in the cited trailing-7-day third-party catalog usage dataset as of 2026-09-06. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share.
Profile data was reviewed for the 2026-09-06 snapshot. Claims retain their source dates, and unavailable fields are shown as unavailable instead of estimated.
Profile facts
- Profile type
- Model research profile
- Provider or publisher
- Alibaba Qwen
- Context window
- 262,144 tokens
- Snapshot date
- 2026-09-06
- Knowledge cutoff
- 2025-06-30
- Reference catalog ID
- qwen/qwen3-235b-a22b-2507
When to pick Qwen3 235B A22B Instruct 2507
Qwen3 235B A22B Instruct 2507 is a reference profile: Alibaba Qwen publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call.
Each condition below is derived from this model's own values in the 2026-09-06 snapshot, compared only against models publishing the same field.
- Reach for it when the model has to call tools. Tool calling is recorded in the snapshot, so this route can drive an agent loop rather than only answer in prose.
- Look elsewhere when you need to call it through Kendr today. This is a research reference profile with no Kendr alias. The facts and evidence here are published for comparison; the model is not routable on Kendr.
Context, modalities, and identity
Qwen3 235B A22B Instruct 2507 accepts text and returns text. The snapshot records structured output, text generation, and tool calling as capabilities or interfaces.
- Provider or publisher
- Alibaba Qwen
- Context window
- 262,144 tokens
- Maximum output
- 16,384 tokens
- Input modalities
- text
- Output modalities
- text
- Knowledge cutoff
- 2025-06-30
Model overview
Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following, logical reasoning, math, code, and tool usage. The model supports a native 262K context length and does not implement "thinking mode" (<think> blocks). Compared to its base variant, this version delivers significant gains in knowledge coverage, long-context reasoning, coding benchmarks, and alignment with open-ended tasks. It is particularly strong on multilingual understanding, math reasoning (e.g., AIME, HMMT), and alignment evaluations like Arena-Hard and WritingBench.
- Reference model ID
- qwen/qwen3-235b-a22b-2507
- Canonical version
- qwen/qwen3-235b-a22b-07-25
- Hugging Face ID
- Qwen/Qwen3-235B-A22B-Instruct-2507
Capabilities and supported controls
Capabilities and parameters are reported from the dated catalog and source records; their presence does not guarantee identical behavior across every provider route.
- Supported parameters
- frequency_penalty, logit_bias, logprobs, max_tokens, min_p, presence_penalty, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p
- structured output
- text generation
- tool calling
Dated pricing snapshot
The dated reference snapshot lists $0.0875 input and $0.35 output per million tokens. Cached input is $0.0175 per million tokens.
These are dated third-party reference prices, not Kendr prices or an availability offer. Routes, tiers, caching, region, and provider policy can change the landed price.
- Price date
- 2026-09-06
- Input
- $0.0875 per 1M tokens
- Cached input
- $0.0175 per 1M tokens
- Output
- $0.35 per 1M tokens
Provider routes, performance, and uptime
The snapshot retains 11 provider endpoints. Provider prices, context limits, p50 performance, and uptime can differ by route and are not Kendr guarantees.
- Active provider endpoints
- 10
- Best p50 latency
- 0.49 s
- Best p50 throughput
- 54 tok/s
- Availability with routing
- 99.82% over the sampled window
- Availability without routing
- 95.19% over the sampled window
- Performance date
- 2026-09-06
| Provider | Quantization | Input / 1M | Output / 1M | Cache read / 1M | Context | p50 latency | p50 throughput | Uptime (1d) |
|---|---|---|---|---|---|---|---|---|
| GMICloud | fp8 | $0.0875 / 1M | $0.35 / 1M | $0.0175 / 1M | 262,144 tokens | 1.25 s | 16 tok/s | 98.91% |
| DeepInfra | fp8 | $0.09 / 1M | $0.55 / 1M | No verified rate | 262,144 tokens | 0.59 s | 9 tok/s | 96.97% |
| NovitaAI | fp8 | $0.09 / 1M | $0.58 / 1M | No verified rate | 131,072 tokens | 0.70 s | 30 tok/s | 99.18% |
| Parasail | fp8 | $0.14 / 1M | $0.8 / 1M | $0.05 / 1M | 131,072 tokens | 0.50 s | 28 tok/s | 99.80% |
| Alibaba Cloud Int. | unknown | $0.1495 / 1M | $0.598 / 1M | No verified rate | 131,072 tokens | 0.49 s | 38 tok/s | 99.99% |
| Venice | fp8 | $0.15 / 1M | $0.75 / 1M | No verified rate | 128,000 tokens | 0.81 s | 12 tok/s | 97.58% |
| AtlasCloud | fp8 | $0.2 / 1M | $0.88 / 1M | $0.2 / 1M | 131,072 tokens | 1.02 s | 27 tok/s | 98.98% |
| StreamLake | unknown | $0.21 / 1M | $0.84 / 1M | No verified rate | 128,000 tokens | 0.64 s | 30 tok/s | 99.10% |
| Google Vertex (US) (ZDR) | unknown | $0.22 / 1M | $0.88 / 1M | No verified rate | 262,144 tokens | 0.60 s | 54 tok/s | 99.94% |
| Google Vertex (US) (ZDR) | unknown | $0.25 / 1M | $1 / 1M | No verified rate | 262,144 tokens | 0.51 s | 34 tok/s | 98.25% |
| Nebius Token Factory | fp8 | $0.2 / 1M | $0.6 / 1M | No verified rate | 262,144 tokens | 0.89 s | 45 tok/s | 66.74% |
Benchmark evidence and limitations
The 2026-09-06 snapshot includes 6 Design Arena categories and AutoExacto results for 13 provider observations. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable.
Third-party catalog fields normalized from the cited model dataset; benchmark methodology and coverage differ by source.
- AutoExacto coverage
- 13 provider observations over 32 days
| Design Arena category | Elo | Win rate | Rank |
|---|---|---|---|
| 3d | 1032 | 41.1% | #99 |
| codecategories | 1055 | 42.6% | #100 |
| dataviz | 1089 | 49% | #92 |
| gamedev | 981 | 35% | #112 |
| uicomponent | 987 | 38.4% | #103 |
| website | 1071 | 43.6% | #100 |
AutoExacto provider benchmarks
Rolling provider observations from the cited third-party benchmark view. The lookback is 32 days; GPQA Diamond and TAU-Bench Airline measure different abilities and should not be blended into one score.
| Provider | GPQA Diamond | TAU-Bench Airline | Runs |
|---|---|---|---|
| Alibaba Cloud Int. | 73.64% | 39.38% | 4 |
| AtlasCloud | 69.38% | 40.17% | 4 |
| auto-routing | 72.78% | 47.32% | 4 |
| Crusoe | 76.33% | 47.92% | 2 |
| DeepInfra | 73.1% | 42.91% | 4 |
| GMICloud | 75.1% | 47.05% | 3 |
| Google Vertex | 73.19% | 45.27% | 4 |
| Google Vertex | 72.31% | — | 4 |
| Nebius Token Factory | 74.48% | 44.56% | 4 |
| NovitaAI | 72.49% | 41.8% | 4 |
| Parasail | 74.04% | 49.09% | 4 |
| StreamLake | 74.48% | 44.27% | 4 |
| Venice | 72.94% | 45.86% | 4 |
Popularity and market context
Qwen3 235B A22B Instruct 2507 ranked #83 in the cited trailing-7-day third-party catalog usage dataset as of 2026-09-06. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share.
- Third-party catalog rank
- #83
- Global market share
- Not inferred
- Observation date
- 2026-09-06
Frequently asked questions
When should I use Qwen3 235B A22B Instruct 2507?
Qwen3 235B A22B Instruct 2507 is a reference profile: Alibaba Qwen publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call. Reach for it when the model has to call tools: Tool calling is recorded in the snapshot, so this route can drive an agent loop rather than only answer in prose. Look elsewhere when you need to call it through Kendr today: This is a research reference profile with no Kendr alias. The facts and evidence here are published for comparison; the model is not routable on Kendr.
What is Qwen3 235B A22B Instruct 2507?
Qwen3 235B A22B Instruct 2507 is a Alibaba Qwen model profile with a 262,144-token context window. The dated catalog snapshot records text modalities and interfaces for structured output, text generation, tool calling.
What context window does Qwen3 235B A22B Instruct 2507 have?
The dated profile lists 262,144 tokens of context and up to 16,384 output tokens. Provider routes, variants, and runtime configuration can impose lower effective limits.
What benchmark evidence is available for Qwen3 235B A22B Instruct 2507?
The 2026-09-06 snapshot includes 6 Design Arena categories and AutoExacto results for 13 provider observations. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable. Third-party catalog fields normalized from the cited model dataset; benchmark methodology and coverage differ by source.
Is Qwen3 235B A22B Instruct 2507 available through Kendr?
This is an independent knowledge profile, not a Kendr-hosted availability claim. Check Kendr's live public model API for currently enabled Kendr aliases.
Does the popularity rank represent Qwen3 235B A22B Instruct 2507's global market share?
Qwen3 235B A22B Instruct 2507 ranked #83 in the cited trailing-7-day third-party catalog usage dataset as of 2026-09-06. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share. No global market-share percentage is inferred when the source does not publish one.
Sources and evidence dates
- Qwen3 235B A22B Instruct 2507 third-party catalog record (catalog, checked 2026-09-06)
- Third-party model catalog methodology (methodology, checked 2026-09-06)
- Alibaba Qwen official model documentation (primary, checked 2026-09-06)
- Design Arena leaderboard and methodology (benchmark, checked 2026-09-06)
- Qwen3 235B A22B Instruct 2507 third-party catalog record (benchmark, checked 2026-09-06)