Skip to main content

Qwen3.5 35B A3B

The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. Its overall performance is comparable to that of the Qwen3.5-27B.

Input / output price
$0.057 / $0.459 per 1M
Context
262K tokens
Released
Feb 24, 2026
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "qwen/qwen3.5-35b-a3b".

Read docs

Providers

7 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.

ProviderInputOutput
Concentrateconcentrate/qwen3.5-35b-a3b
$0.057/M$0.459/M
EmpirioLabsempiriolabs/qwen3.5-35b-a3b
$0.057/M$0.459/M
DeepInfradeepinfra/qwen3.5-35b-a3b
$0.14/M$1.00/M
Parasailparasail/qwen3.5-35b-a3b
$0.15/M$1.00/M
Novitanovita/qwen3.5-35b-a3b
$0.25/M$2.00/M

Is a provider down? OpenRouter status

Pricing

Averaged across 7 providers, in USD per 1M tokens after gateway discounts.

Average input
$0.15 per 1M
Average output
$1.13 per 1M
Average cache read
$0.05 per 1M

Cost by volume

Total cost from 0 to 1T tokens on the cheapest and most expensive provider.

Highest$0.25/MNovita and QwenLowest$0.057/MConcentrate and EmpirioLabs

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

52tokens/sacross providers

Faster than 66% of the other 596 models on the gateway

Response latency

Lower is better

3.7sacross providers

Faster than 36% of the other 758 models on the gateway

Benchmarks

The Respan Index plus 7 published scores from 4 sources, each linked to where it was published.

18.4/ 100

Respan Index#144 of 176

176 ranked models. The shaded band is this model's 95% interval, 13.6 to 29.8. Qwen3 235B-A22B Thinking 2507 and o3-mini fall inside it, so the gap is within the margin of error.
Qwen3.5 35B A3B by category: Coding 0.9, Agents 15.5 (estimated), Reasoning 23.4, Knowledge & language 13.9 (estimated), Human preference 47.1
  1. 144Qwen3.5 35B-A3B18.4
  2. 142Qwen3 235B-A22B Thinking 250718.5
  3. 145o3-mini17.7

Published scores

BenchmarkScore
SWE-rebench 2026-05-15 to 2026-07-01Nebius17.1%

Compare Qwen3.5 35B A3B: Qwen3.5 35B-A3B vs Qwen3.6 35B-A3B, Qwen3.5 35B-A3B vs Qwen3.8-Max, Qwen3.5 35B-A3B vs Qwen3.8-Flash-Next, Qwen3.5 35B-A3B vs Qwen3.7-Max, Qwen3.5 35B-A3B vs Qwen3.8 27B, Qwen3.5 35B-A3B vs Qwen3.6-Max-Preview

Apps

Use Qwen3.5 35B A3B in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

Qwen3.5 35B A3B FAQ

Short answers, built from the data on this page.

Qwen3.5 35B A3B is a model from Qwen, released on Feb 24, 2026. The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. Its overall performance is comparable to that of the Qwen3.5-27B.

On the Respan gateway, Qwen3.5 35B A3B costs from $0.057 per 1M input tokens and $0.459 per 1M output tokens on Concentrate and EmpirioLabs, after gateway discounts. Averaged across its 7 providers, it costs $0.15 per 1M input tokens and $1.13 per 1M output tokens.

Qwen3.5 35B A3B is available through 7 providers: Concentrate, EmpirioLabs, DeepInfra, Parasail, Novita, Qwen, and 1 more.

Qwen3.5 35B A3B accepts up to 262K tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "qwen/qwen3.5-35b-a3b". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Qwen3.5 35B A3B scores 18.4 out of 100 on the Respan Index v1.1, ranking #144 of 176 models. By category: Coding 0.9, Agents 15.5 (estimated), Reasoning 23.4, Knowledge & language 13.9 (estimated), and Human preference 47.1. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

Qwen3.5 35B A3B has 7 published benchmark scores, including SWE-rebench 2026-05-15 to 2026-07-01 17.1%, GPQA Diamond 83.5%, and AIME 2026 93.3%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.