Migrating from qwen3-235b-a22b-instruct-2507 to qwen3.5-397b-a17b
What changes when you move from qwen3-235b-a22b-instruct-2507 to qwen3.5-397b-a17b: input price +160.9%, output price +291.3%.
- Status
- Retired
- Shutdown
- Oct 10, 2026
- Time left
- —
- Status
- Active
- Shutdown
- none announced
- Time left
- —
Prices
USD per 1M tokens, standard tier| Type | qwen3-235b-a22b-instruct-2507 | qwen3.5-397b-a17b | Change |
|---|---|---|---|
| Input | $0.23first seen October 5, 2026 · source | $0.60first seen October 5, 2026 · source | +160.9% |
| Output | $0.92first seen October 5, 2026 · source | $3.60first seen October 5, 2026 · source | +291.3% |
At 1M input and 250K output tokens a month: $0.46 on qwen3-235b-a22b-instruct-2507, $1.50 on qwen3.5-397b-a17b (+226.1%).
Other tiers, prompt-size bands and modalities are on each model's page.
Limits
Tokens| Limit | qwen3-235b-a22b-instruct-2507 | qwen3.5-397b-a17b | Change |
|---|---|---|---|
| Context window | Not published | 262,144 tokensin place when tracking began (October 6, 2026) · source | — |
| Max output tokens | Not published | 65,536 tokensin place when tracking began (October 6, 2026) · source | — |
Benchmarks
Same benchmark, source and setting on both sides| Benchmark | qwen3-235b-a22b-instruct-2507 | qwen3.5-397b-a17b | Change |
|---|---|---|---|
| Arena TextArena. Ranges don't overlap. | 1422 ±3 | 1442 ±3 | +20 |
| Arena Text: CodingArena. Ranges don't overlap. | 1472 ±5 | 1492 ±5 | +20 |
A score from one source is never compared with another source's, and company-reported scores are compared only within the same company; settings named where they differ. Data: Arena (arena.ai), CC BY 4.0.
Features and parameters
Compare modelsNo feature is published for both models, so none can be compared.
Not published for qwen3-235b-a22b-instruct-2507: Structured outputs, Function calling, Prompt caching, Batch API, Image input, PDF input, Reasoning / thinking, Fine-tuning, Built-in web search, temperature parameter, top_p parameter, logprobs, Context window, Max output tokens.
Not published for qwen3.5-397b-a17b: PDF input, Reasoning / thinking, temperature parameter, top_p parameter, logprobs.
Aliases
All aliasesNo aliases point at either model.
On cloud platforms
All platform dates| Model | Platform | Platform ID | Shutdown | Time left |
|---|---|---|---|---|
| qwen3-235b-a22b-instruct-2507 | Amazon Bedrock | qwen.qwen3-235b-a22b-2507-v1:0 | not before Apr 28, 2026Active | |
| qwen3-235b-a22b-instruct-2507 | Google Cloud Vertex AI | qwen3-235b-a22b-instruct-2507-maas | Oct 21, 2026Deprecated | 6 days |
Read daily from each platform's own pages. The status above uses each company's own date. Time left counts to 00:00 UTC on the shutdown date, which is the evening before in ET.
Checklist
- Change the model string from
qwen3-235b-a22b-instruct-2507toqwen3.5-397b-a17b. - Re-check costs: qwen3.5-397b-a17b costs more than qwen3-235b-a22b-instruct-2507 (input +160.9%, output +291.3%).
Compared from AIWire's published data when this page loaded. Each fact links to the page it came from. Model pages: qwen3-235b-a22b-instruct-2507, qwen3.5-397b-a17b.