Skip to content
ModelCap

Head-to-head comparison

Qwen3 235B A22B Instruct 2507 vs Qwen3.5 397B A17B

Qwen3 235B A22B Instruct 2507 (Qwen) and Qwen3.5 397B A17B (Qwen) compared on the ModelCap Index, API price, context window, provider availability, weight access and every public benchmark board they share. Figures are the same ones shown on the live rankings; nothing here is a hidden score.

Snapshot as of 22 September 2026

As of 22 September 2026, Qwen3.5 397B A17B holds the stronger ModelCap Index position (#42 vs #60); and Qwen3 235B A22B Instruct 2507 is 10× cheaper per output token ($0.35/1M vs $3.50/1M).

Which should you choose?

Choose Qwen3 235B A22B Instruct 2507 if…

  • API cost matters — $0.35/1M output tokens against $3.50/1M, about 10× cheaper.
  • your workload is prompt-heavy — input tokens cost $0.087/1M against $0.55/1M.

Choose Qwen3.5 397B A17B if…

  • you want the stronger overall ModelCap Index position — #42 against #60 (69.7 vs 63.6 points).
  • you want provider choice — 10 listed API providers against 9.
  • Arena coding is your yardstick — 1491 against 1472.
  • AA τ²-Bench Telecom is your yardstick — 95.6% against 33.3%.
  • AA Humanity's Last Exam is your yardstick — 29.0% against 11.1%.
  • you want the more recently listed model — Qwen3.5 397B A17B was listed 16 February 2026, Qwen3 235B A22B Instruct 2507 21 July 2025.

Qwen3 235B A22B Instruct 2507 vs Qwen3.5 397B A17B: specs, pricing and context

Specification comparison of Qwen3 235B A22B Instruct 2507 and Qwen3.5 397B A17B
FieldQwen3 235B A22B Instruct 2507QwenQwen3.5 397B A17BQwen
ModelCap Index position#60#42
Index score63.669.7
EvidenceMeasuredMeasured
Input price / 1M tokens$0.087$0.55
Output price / 1M tokens$0.35$3.50
Context window262K tokens262K tokens
Max output tokens236K236K
API providers listed910
Weight accessOpen weightsOpen weights
Input modalitiesTextText, Image, Video
First listed21 July 202516 February 2026
PublisherQwenQwen

Evidence: 2 public benchmark observations across 2 boards · 3 public benchmark observations across 3 boards. Prices are the lowest listed API offer per million tokens observed on the OpenRouter catalogue.

Benchmark scores: Qwen3 235B A22B Instruct 2507 vs Qwen3.5 397B A17B

Public benchmark boards where Qwen3 235B A22B Instruct 2507 or Qwen3.5 397B A17B has a published result
BoardQwen3 235B A22B Instruct 2507Qwen3.5 397B A17BLeads
Arena codingArena (LMArena)14721468–147714911486–1496Qwen3.5 397B A17B
BFCL V4UC Berkeley (Gorilla)52.2Prompt—Only one result
AA Intelligence IndexArtificial Analysis—18.4intelligence-index:v4.3:reasoning-unspecifiedOnly one result
AA τ²-Bench TelecomArtificial Analysis33.3%tau2:telecom:dual-control:pass-at-1:3-repeats:source-model="Qwen3 235B A22B 2507 Instruct":reasoning=false95.6%tau2:telecom:dual-control:pass-at-1:3-repeats:source-model="Qwen3.5 397B A17B (Reasoning)":reasoning=trueQwen3.5 397B A17B
AA Humanity's Last ExamArtificial Analysis11.1%hle:may-2025:text-only-2158:no-tools:pass-at-1:source-model="Qwen3 235B A22B 2507 Instruct":reasoning=false29.0%hle:may-2025:text-only-2158:no-tools:pass-at-1:source-model="Qwen3.5 397B A17B (Reasoning)":reasoning=trueQwen3.5 397B A17B

Scores are the sources' own published figures for each model's best evaluated configuration; ModelCap never re-runs a benchmark.

Want a different pairing? Open the interactive comparison tool to swap either model for any current ranked language model.

Qwen3 235B A22B Instruct 2507 vs Qwen3.5 397B A17B: common questions

Is Qwen3 235B A22B Instruct 2507 better than Qwen3.5 397B A17B?

Qwen3.5 397B A17B ranks higher on the ModelCap Index as of 22 September 2026: #42 against #60. That is a capability ranking built from public benchmark evidence with published uncertainty; whether it is "better" for you also depends on price, context and where you can run it. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.

Is Qwen3 235B A22B Instruct 2507 cheaper than Qwen3.5 397B A17B?

Qwen3 235B A22B Instruct 2507 is cheaper on output tokens: $0.35/1M against $3.50/1M. Input tokens are $0.087/1M for Qwen3 235B A22B Instruct 2507 and $0.55/1M for Qwen3.5 397B A17B. Prices are the lowest listed API offer ModelCap observed, in USD per million tokens. For 1,000 requests with 2,000 input and 500 output tokens each (2M input + 0.5M output), the listed-rate estimate is $0.35 for Qwen3 235B A22B Instruct 2507 versus $2.85 for Qwen3.5 397B A17B. Qwen3 235B A22B Instruct 2507 costs 87.7% less in this scenario. This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the selected endpoint before budgeting.

Which has the bigger context window, Qwen3 235B A22B Instruct 2507 or Qwen3.5 397B A17B?

Both publish a 262K-token context window.

Which is better for coding, Qwen3 235B A22B Instruct 2507 or Qwen3.5 397B A17B?

Arena coding: Qwen3 235B A22B Instruct 2507 1472, Qwen3.5 397B A17B 1491 — Qwen3.5 397B A17B leads.

Are Qwen3 235B A22B Instruct 2507 and Qwen3.5 397B A17B open-weight models?

Qwen3 235B A22B Instruct 2507: Open weights. Qwen3.5 397B A17B: Open weights. Open weights mean the checkpoint can be downloaded and self-hosted under its licence; API-only models are available solely through hosted endpoints.

Where can I run Qwen3 235B A22B Instruct 2507 and Qwen3.5 397B A17B?

ModelCap currently lists 9 API providers for Qwen3 235B A22B Instruct 2507 and 10 for Qwen3.5 397B A17B, from the OpenRouter catalogue snapshot the site serves; each model page lists the providers and their prices.

How current is this Qwen3 235B A22B Instruct 2507 vs Qwen3.5 397B A17B comparison?

Every figure comes from the sealed ModelCap dataset published 22 September 2026; the page re-renders within a minute of each data refresh, and the ModelCap Index positions are the same ones shown on the live rankings.

Explore further