Skip to content
Alpha
  • Models
  • Rankings
  • Ori

Footer

OpenRouterOpenRouter
© 2026 OpenRouter, Inc

Product

  • Chat
  • Rankings
  • Benchmarks
  • Apps
  • Discover
  • Models
  • Ori
  • Collections
  • Providers
  • Tools
  • Pricing
  • Business
  • Enterprise
  • Labs

Company

  • About
  • Blog
  • Customers
  • Careers
    Hiring
  • Privacy
  • Terms of Service
  • Trust Center
  • Support
  • Works With OR
  • Data
  • Brand

Developer

  • Documentation
  • API Reference
  • Developer Platform
  • Status
  • AI Site Map

Connect

  • Discord
  • GitHub
  • LinkedIn
  • X
  • YouTube
Favicon for qwen

Qwen: Qwen3.5-122B-A10B

qwen/qwen3.5-122b-a10b

Model weights
Compare

The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of overall performance, this model is second only to Qwen3.5-397B-A17B. Its text capabilities significantly outperform those of Qwen3-235B-2507, and its visual capabilities surpass those of Qwen3-VL-235B.

Modalities
In / Out Price
$0.26 / $2.08per 1M
Context
262K
Released
Feb 25, 2026
Compare
ProvidersPricingPerformanceUptimeBenchmarksAppsActivityFAQExplore

Providers

This model is hosted by one provider. OpenRouter forwards every request to it directly — no routing decisions to make.

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Qwen: Qwen3.5-122B-A10B (Artificial Analysis)
SourceBenchmarkScore
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) Intelligence Index17.7
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) Coding Index43.3
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) GPQA Diamond82.7%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) HLE15.9%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) IFBench50.8%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) τ²-Bench Telecom84.5%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) AA-LCR61.3%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) τ-Bench Banking10.3%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) GDPval-AA10.4%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) CritPt0.9%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) Terminal-Bench Hard29.5%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) Terminal-Bench 2.147.2%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) AA-Omniscience Accuracy19.1%
Artificial AnalysisQwen3.5 122B A10B (Non-reasoning) AA-Omniscience Non-Hallucination Rate8.4%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) Intelligence Index15.6
Artificial AnalysisQwen3.5 122B A10B (Reasoning) Coding Index45.7
Artificial AnalysisQwen3.5 122B A10B (Reasoning) Agentic Index7.6
Artificial AnalysisQwen3.5 122B A10B (Reasoning) GPQA Diamond85.7%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) HLE25.2%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) IFBench75.7%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) τ²-Bench Telecom93.6%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) AA-LCR76.3%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) τ-Bench Banking15.3%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) GDPval-AA15.7%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) CritPt0.6%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) SciCode39.7%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) Terminal-Bench Hard31.1%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) Terminal-Bench 4.00.0%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) Terminal-Bench 2.147.6%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) AA-Omniscience Accuracy24.4%
Artificial AnalysisQwen3.5 122B A10B (Reasoning) AA-Omniscience Non-Hallucination Rate12.9%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Explore more models

AI Models with Vision: Multimodal LLMs for Image UnderstandingCollectionAI Model RankingsRanking
Latency / throughput
$0.26$2.08--1.19s53 tps
96.54%
$0.26$2.08--0.94s84 tps
99.89%
$0.30$2.40$0.302.19s66 tps
96.22%
$0.40$3.20--2.27s58 tps
99.95%
Throughput
84tok/s

P50, best across providers

Latency
0.94s

P50, best provider

AutoExacto benchmarks
GPQA DiamondTAU-BenchSiliconFlow84.7%--DeepInfra83.1%--AtlasCloud84.5%79.7%auto-routing84.9%78.7%NovitaAI84.5%78.3%
+1 more providers
Uptime (3d)The model was reachable. Request routed to a provider.
100.00%
Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.
99.72%

Availability over the last 3 days

Last 72 hours
Availability 99.72%
3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability
99.68%
Without Routing
97.29%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

1.
Favicon for https://nousresearch.com
Hermes Agent
Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.
933Mtokens
2.
Favicon for https://pi.dev/
pi
There are many coding agents, but this one is yours.
680Mtokens
3.
Favicon for https://github.com/deepseek-ai/deepseek-harness
DeepSeek Harness
new
642Mtokens
4.
Favicon for https://github.com/z4-qqq/materials-design-agents-v2
Materials Design Agents
new
333Mtokens
5.
Favicon for https://openai.com/codex
Codex
A coding agent that helps you build and ship with AI
249Mtokens

Frequently asked questions

The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of overall performance, this model is second only to Qwen3.5-397B-A17B.

Qwen3.5-122B-A10B costs $0.26/M input tokens and $2.08/M output tokens.

Qwen3.5-122B-A10B has a 262,144 token context window. It supports up to 262,144 completion tokens.

No. Qwen3.5-122B-A10B does not accept tools, so function calling is unavailable. It also supports structured outputs via a JSON schema in response_format.

Qwen3.5-122B-A10B accepts text, images, and video as input and returns text.

Qwen3.8 Max Prime, Qwen3.8 Omni Flash, Qwen3.8 Max (0902) and 38 more are other text models from Qwen.

Qwen3.5-122B-A10B was released on February 25, 2026.