LLM Pricing
PricingLeaderboardsToolsProvidersReleasesGuides

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Qwen3 4B

alibaba·alibaba/qwen3-4b-fp8·GA·Open weights
Qwen instruction model for multilingual chat, reasoning, and tool use

Specs & pricing

Input / output per 1M tokens
Reference price·NovitaAI
$0.03 / $0.03
Blended $0.03 · Cache read —
Lowest paid·NovitaAIGateway
$0.03 / $0.03
Blended $0.03
Context
128,000
Output limit
20,000
Knowledge cutoff
—
Released / updated
2025-04-29 / 2025-04-29
Capabilities
✓ ReasoningTool use? Structured output✓ TemperatureAttachments
Modalities
Text

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

Intelligence7.2
Value240
Coding—
Agentic—
Output speed—
TTFT—
Value formulaIQ 7.2 ÷ min blended $0.030 = 240
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
Non-reasoningIQ 6.6
ReasoningIQ 7.2

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

Available at 1 providers1 with public prices

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
NovitaAIGateway$0.03$0.03——128,00020,000

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

Upstream provides no control info

1 / 1 providers expose no reasoning control (reasoning_options: []).

Your usage cost

1NovitaAI$7.50
The cheapest paid channel is the only channel.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$0.03$02026-08-052026-08-132026-08-05 · Input list $0.032026-08-06 · Input list $0.032026-08-07 · Input list $0.032026-08-08 · Input list $0.032026-08-09 · Input list $0.032026-08-10 · Input list $0.032026-08-11 · Input list $0.032026-08-12 · Input list $0.032026-08-13 · Input list $0.032026-08-05 · Output list $0.032026-08-06 · Output list $0.032026-08-07 · Output list $0.032026-08-08 · Output list $0.032026-08-09 · Output list $0.032026-08-10 · Output list $0.032026-08-11 · Output list $0.032026-08-12 · Output list $0.032026-08-13 · Output list $0.032026-08-05 · Min blended $0.032026-08-06 · Min blended $0.032026-08-07 · Min blended $0.032026-08-08 · Min blended $0.032026-08-09 · Min blended $0.032026-08-10 · Min blended $0.032026-08-11 · Min blended $0.032026-08-12 · Min blended $0.032026-08-13 · Min blended $0.03

Artificial Analysis evaluations10 items

GPQA Diamond52.2%
Humanity's Last Exam4.4%
MMLU-Pro69.6%
LiveCodeBench46.5%
AIME 202522.3%
AIME65.7%
MATH-50093.3%
τ²-Bench Telecom19.0%
AA-LCR0.0%
IFBench32.5%

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

Hy3cheaper alternative$0 / $0Nemotron 3.5 Lightning 30B A3Bcheaper alternative$0 / $0Laguna S 2.1cheaper alternative$0 / $0GLM-4.6V-Flashcheaper alternative$0 / $0
Data partly from models.dev (MIT) · NovitaAI official docs ↗