LLM Pricing
PricingLeaderboardsToolsProvidersReleasesGuides

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Qwen3.5 4B

alibaba·alibaba/qwen3-5-4b·GA·Open weights·qwen series
Qwen3.5 4B is a low-cost multimodal reasoning model with 256K context, image and video input, function tools, and structured output.

Qwen3.5 4B by alibaba is offered by 2 providers on this page. Public prices are shown for 1 of them. Its reference price is $0.04 per 1M input tokens and $0.07 per 1M output tokens. It also has 1 free ($0) channel; free tiers usually carry rate limits, and subscription-covered access bills $0 per token only after the subscription fee.

Artificial Analysis rates it 13.1 on the Intelligence Index, with 22.6 for coding. Against its lowest blended price of $0.048 per 1M, that is roughly 276 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 26 tokens per second, with 0.72s to the first token.

The context window is 262,144 tokens at the reference host, but hosts report different limits, from 32,768 to 262,144, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, and Video. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use.

Specs & pricing

Input / output per 1M tokens
Reference price·EmpirioLabs AI
$0.04 / $0.07
Blended $0.048 · Cache read $0.02
Lowest paid·EmpirioLabs AIGateway
$0.04 / $0.07
Blended $0.048
1 more $0 channels
Context
262,144
Output limit
32,768
Knowledge cutoff
—
Released / updated
2026-03-02 / 2026-03-02
Capabilities
✓ Reasoning✓ Tool use✓ Structured output✓ Temperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImageVideo

Quality & performanceArtificial Analysis · Intelligence Index v4.3 · rep. tier Reasoning

Intelligence13.1
Value276
Coding22.6
Agentic—
Output speed26 tok/s
TTFT0.72 s
Value formulaIQ 13.1 ÷ min blended $0.048 = 276
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
Non-reasoningIQ 10.8 · 23 tok/s
ReasoningIQ 13.1 · 26 tok/s

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 2 providers1 with public prices · 1 free

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
QVAC
qwen3.5-4b
GatewayFree——32,768 ⚠8,192
EmpirioLabs AIGateway$0.04$0.07$0.02—262,14432,768

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Reasoning control

effort = noneeffort = loweffort = mediumeffort = higheffort = maxbudget_tokens ≥ 1,024 ≤ 32,768

Your usage cost

1EmpirioLabs AI$9.10

1 more channels offer $0 (QVAC); free tiers usually have rate limits and no SLA, excluded from ranking.

The cheapest paid channel is the only channel.

Price historyone sample accumulated per data sync

Input list $0.04Output list $0.07Min blended $0.048

Price history accumulates from each data sync; currently only 1 sample(s) (2026-09-24). Each future sync adds a point, and once accumulated a line is drawn here.

Artificial Analysis evaluations8 items

GPQA Diamond77.1%
Humanity's Last Exam9.9%
Terminal-Bench Hard18.2%
Terminal-Bench 2.125.8%
τ²-Bench Telecom92.1%
τ³-Bench Banking6.8%
AA-LCR63.0%
IFBench52.0%

Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and harness, so rows are not comparable with one another. The Intelligence Index above draws on a different, newer set of evaluations.

Related models

Qwen 3.8 27B Hemingwaysame series$0.25 / $1.50Qwen 3.8 27B Cybersecuritysame series$0.10 / $0.60Qwen3.8 Omni Flashsame series$0.30 / $0.94Hy3cheaper alternative$0 / $0Nemotron 3.5 Lightning 30B A3Bcheaper alternative$0 / $0Nemotron 3 Nano 30B A3Bcheaper alternative$0 / $0Laguna S 2.1cheaper alternative$0 / $0
Data partly from models.dev (MIT) · EmpirioLabs AI official docs ↗