Migrating from qwen3-30b-a3b-instruct-2507 to qwen3.5-35b-a3b

What changes when you move from qwen3-30b-a3b-instruct-2507 to qwen3.5-35b-a3b: input price +25%, output price +150%.

From: qwen3-30b-a3b-instruct-2507
Status: Retired
Shutdown: Oct 10, 2026
Time left: —

To: qwen3.5-35b-a3b
Status: Active
Shutdown: none announced
Time left: —

Prices

USD per 1M tokens, standard tier

Type	qwen3-30b-a3b-instruct-2507	qwen3.5-35b-a3b	Change
Input	$0.20first seen October 5, 2026 · source	$0.25first seen October 5, 2026 · source	+25%
Output	$0.80first seen October 5, 2026 · source	$2.00first seen October 5, 2026 · source	+150%

At 1M input and 250K output tokens a month: $0.40 on qwen3-30b-a3b-instruct-2507, $0.75 on qwen3.5-35b-a3b (+87.5%).

Other tiers, prompt-size bands and modalities are on each model's page.

Limits

Tokens

Limit	qwen3-30b-a3b-instruct-2507	qwen3.5-35b-a3b	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 and source on both sides; settings named where they differ

Benchmark	qwen3-30b-a3b-instruct-2507	qwen3.5-35b-a3b	Change
Arena TextArena. Ranges don't overlap.	1382 ±5	1394 ±4	+12
Arena Text: CodingArena.	1437 ±9	1435 ±7	−2 within range
GPQA DiamondEpoch AI. Other settings not run → 81.2% no reasoning	55.6%	83.5%	+27.9 pts
OTIS Mock AIME 2024-2025Epoch AI, no reasoning (vs no setting). Other settings 62.2% → 54.4%	62.2%	70.0%	+7.8 pts

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; Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.

Features and parameters

Compare models

No feature is published for both models, so none can be compared.

Not published for qwen3-30b-a3b-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 qwen3.5-35b-a3b: PDF input, Reasoning / thinking, temperature parameter, top_p parameter, logprobs.

Aliases

All aliases

No aliases point at either model.

Checklist

Change the model string from qwen3-30b-a3b-instruct-2507 to qwen3.5-35b-a3b.
Re-check costs: qwen3.5-35b-a3b costs more than qwen3-30b-a3b-instruct-2507 (input +25%, output +150%).

Compared from AIWire's published data when this page loaded. Each fact links to the page it came from. Model pages: qwen3-30b-a3b-instruct-2507, qwen3.5-35b-a3b.