Qwen3.5 4B
Qwen3.5 4BAlternatives?
CheaperQwen3.7 Flashunstated+55%-8%$0.03 / $0.13
BetterGPT-5.6 Lunalow+19%+650%$0.01OpenQwen 3.5 9Bnone+10%+54%$0.08 / $0.13Memory (4-bit): 2.3 GB → 4.8 GBCheaperGPT OSS 20BmediumOpen+17%-40%$0.018 / $0.09Context: 262K → 131K+2
BetterGemini 3.1 Flash Litehigh+43%+838%$0.06Qwen3.7 Flashunstated+55%-8%$0.03 / $0.13GPT OSS 20BmediumOpen+17%-40%$0.018 / $0.09Context: 262K → 131K+2GPT-5.6 Lunalow+19%+650%$0.01Gemini 3.1 Flash Litehigh+43%+838%$0.06GPT-6 Lunamax+77%+233%$0.068Function calling only with none · Chat Completions+3
Claude Haiku 5.5high+74%+233%$0.079Remove sampling parameters+5GPT OSS 120BhighOpen+59%same$0.092Context: 262K → 131K+2Qwen 3.5 9BnoneOpen+10%+54%$0.08 / $0.13Memory (4-bit): 2.3 GB → 4.8 GBQwen3 8BunstatedOpensame+417%$0.18 / $0.70Context: 262K → 33K+2Qwen3 4B 2507 InstructunstatedOpen-6%+733%$0.40 / $0.80No image input
DeepSeek R1 0528 Qwen3 8BunstatedOpen-21%+13%$0.06 / $0.09Context: 262K → 33K+2Gemma 3 12BunstatedOpen-70%+4%$0.05 / $0.10Context: 262K → 131K+1Qwen3.8 Max (0902)xhigh+79%—$2 / $6Qwen3.7 Maxmax+71%—$1.48 / $4.43No image inputQwen3.6 Plusunstated+67%—$0.50 / $3Qwen3.6 Max Previewunstated+63%—$1.30 / $7.80No image input+1Qwen 3.5 Plusunstated+55%—$0.12 / $0.69Qwen3 235B A22B Thinking 2507unstatedOpen+55%—$0.23 / $2.30Memory (4-bit): 2.3 GB → 141 GB+2Qwen3.6 Flashunstated+51%—$0.25 / $1.50Qwen 3.5 Flashunstated+51%—$0.065 / $0.26All Alibaba (Qwen) models →Run it yourself
estimateFAQ
How much does Qwen3.5 4B cost?›
$0.03 per 1M input tokens and $0.15 per 1M output tokens (cheapest host: DeepInfra).
Is Qwen3.5 4B free to use?›
There is no free API; the cheapest paid price is $0.03 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Qwen3.5 4B, and what is it best at?›
#192 of 334 in Chat & Knowledge, #95 of 239 in Coding, #36 of 242 in Agents & Tool Use, #110 of 245 in Writing & Language, #79 of 124 in Vision & Documents. Strongest at: Tool calling.
Is Qwen3.5 4B better than Qwen3 4B 2507 Instruct?›
On OTIS Mock AIME 2024 2025, Qwen3.5 4B scores 55.8 vs 52.2 — higher.
How can I use Qwen3.5 4B?›
Through the API of DeepInfra, Featherless.
What is the context window of Qwen3.5 4B?›
262K tokens.
When was Qwen3.5 4B released? Is there a newer version?›
Released 2026-03-02. It is the current version in its line.
Is Qwen3.5 4B open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Qwen3.5 4B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Qwen3.5 4B read images?›
It accepts text and images as input.
What is Qwen3.5 4B?›
Qwen3.5 4B is an open-weights language model from Alibaba (Qwen), released Mar 2026, the successor to Qwen3 4B 2507 Instruct.
What are the best alternatives to Qwen3.5 4B?›
Qwen3.7 Flash (cheaper), GPT-5.6 Luna (better), Qwen 3.5 9B (open weights).