Ask
Packaging

Qwen 3.8-27B joins Groq's self-serve catalog with a published price

Groq pricing

Groq's docs added a price for Qwen/Qwen3.8-27B in the Preview Models table — $0.80 input / $4.00 output per 1M tokens at 450 T/SEC — a model that previously had rate limits but no listed price.

Before

qwen/qwen3.8-27b appeared only on the GroqCloud Rate Limits page (30 RPM / 8K TPM / 2M TPD on the Developer plan) with no price and no entry in the models catalog table.

After

Qwen/Qwen3.8-27B listed in the Preview Models pricing table at $0.80 input / $4.00 output per 1M tokens, 450 T/SEC, 131,042-token context window, 16,384 max completion tokens, 20 MB max file size.

A single-model, single-day change: Groq’s GroqCloud docs model catalog (console.groq.com/docs/models) added a published price for Qwen/Qwen3.8-27B in the Preview Models section. The model ID had already been visible on the Rate Limits docs page for at least a day with throughput and quota figures but no price and no row in the models catalog table, so this capture is the first to resolve it to a concrete rate. Every other price, model, and page on Groq’s tracked pricing surfaces (homepage, models catalog, rate limits) matched the prior day’s capture exactly — this was the only change.

From Groq's pricing timeline
Qwen 3.8-27B Gains a Published Preview-Tier Price

Qwen/Qwen3.8-27B joined the GroqCloud docs' Preview Models pricing table at $0.80 input / $4.00 output per 1M tokens (450 T/SEC, 131,042-token context window, 16,384 max completion tokens, 20 MB max file size). The model ID had appeared on the Rate Limits docs page since at least 2026-08-26 (30 RPM / 8K TPM / 2M TPD on the Developer plan) but carried no price and was absent from the models catalog table entirely — this capture is the first to show a published rate. It was the only change versus the prior day's capture: the homepage, every other model price, and the Rate Limits page all held exactly.

About Groq
groq.com ↗

Groq runs a pure-usage per-token serverless inference API on its proprietary LPU silicon. As of 2026-08-11 the dedicated marketing pricing page (groq.com/pricing/) has been removed and redirects to the homepage; rates survive only in the GroqCloud developer docs. As of 2026-08-26, Llama 3.1 8B Instant and Llama 3.3 70B Versatile — previously $0.05/$0.08 and $0.59/$0.79 per 1M tokens — moved to Enterprise-only "Contact Sales" pricing and dropped off both the Free and Developer rate-limit tables. The self-serve catalog is now GPT OSS 20B and Safety GPT OSS 20B (formerly "GPT OSS Safeguard 20B") at $0.075/$0.30 (1,000 T/SEC), GPT OSS 120B at $0.15/$0.60 (500 T/SEC), Qwen 3.6 27B at $0.60/$3.00, and two Preview moderation models — Llama Prompt Guard 2 22M and Prompt Guard 2 86M — at $0.03/$0.03 and $0.04/$0.04 per 1M tokens. As of 2026-08-27, Qwen 3.8-27B joined the Preview catalog with a published price of $0.80/$4.00 per 1M tokens (450 T/SEC) — the only change versus the prior day's capture.

Free tier
Yes
Commits
Available
Transparency
public

Groq pricing history

  1. Aug 2026
    Qwen 3.8-27B Gains a Published Preview-Tier Price
  2. Aug 2026
    Llama 3.1 8B Instant and Llama 3.3 70B Versatile Moved to Enterprise-Only Pricing
  3. Aug 2026
    Public Pricing Page Removed — Rates Survive Only in Developer Docs
  4. Jul 2026
    Catalog Contraction: Two Models and the Browser Automation Tool Withdrawn
  5. Jul 2026
    Text-to-Speech SKU + Browser Automation Tool + New Models
Full Groq timeline

More Groq activity

All pricing activity