> AI21 Labs
AI21 Labs sells API access to its Jamba foundation models through AI21 Studio on usage-based pricing, with a self-serve free trial and a custom plan for volume. The free trial is published as $10 in credits for 7 days with no credit card required. Published pay-as-you-go rates are Jamba Mini at $0.2 per 1M input tokens and $0.4 per 1M output tokens, and Jamba Large at $2 per 1M input tokens and $8 per 1M output tokens. AI21 states that on Studio models an average token equals about 1 word or 6 characters of English text.
// pros
- The free trial is stated in dollar terms rather than tokens: $10 of credits for 7 days, no credit card required, which makes it easy to price a specific evaluation before committing.
- Jamba Mini is the cheapest published tier at $0.2 per 1M input tokens and $0.4 per 1M output tokens, roughly a tenth of Jamba Large input pricing.
- The pay-as-you-go plan has no seat minimum and includes foundation model APIs and SDK access, so a single developer key can be used across a team.
- AI21 publishes a token-density claim for Studio models (an average token equals about 6 characters of English, described as up to 30% more text per token than other providers), which is directly checkable against your own corpus before switching.
- The custom plan is the only tier that lists premium API rate limits, private cloud hosting, volume discounts and a dedicated account manager.
// cons
- The $10 trial credit expires after 7 days and there is no perpetual free allowance published on the pricing page.
- Rate limits are not published numerically: premium API rate limits appear only as a bullet on the Custom plan, so default throughput cannot be sized from public documentation.
- Jamba Large output is priced at $8 per 1M tokens, twenty times the Jamba Mini output rate, so long generations on the large model dominate cost.
- Only two model tiers are listed on the public pricing page, which limits the options for matching price to task without a sales conversation.
Spend the 7-day, $10 trial on one bounded evaluation script so the credits are not lost, and start production on Jamba Mini for drafting, extraction and classification. Reserve Jamba Large for long-context reasoning passes where the $2 per 1M input token price is justified, and only open a Custom plan conversation once a measured rate limit, not a hypothetical one, is the blocker.
written by Mike Fleming, Uprouter editorial · verified against official sources September 16, 2026
Value taken from the provider’s own published free-tier terms, expressed in USD.
note: Official pricing page states a Free Trial of $10 credits for 7 days, no credit card needed (ai21.com/pricing, checked 2026-09-16).
| Plan | Type | Monthly | Input /1M | Output /1M | Note |
|---|---|---|---|---|---|
| Free tier (source-reported) | free | — | — | — | New accounts receive a $10 trial credit valid for 3 months; once exhausted or expired you must add billing info to continue. Trial rate limits are generous (~200 RPM / 10 RPS), no credit card required. |
| Jamba 1.5 Mini | payg | — | $0.050 | $0.200 | High-efficiency SSM-Transformer hybrid model. |
| Jamba 1.5 Large | payg | — | $2.00 | $8.00 | — |
Pricing normalized from public sources — always verify with the provider.
// model prices
| Model | Input /1M | Output /1M | Context |
|---|---|---|---|
| Jamba Large (1.7) | — | — | — |
| Jamba Mini (1.7) | — | — | — |
| Jamba 1.6 Large | — | — | — |
$ Does AI21 Labs have a free tier?
Yes — AI21 Labs offers a free tier that we value at roughly $10.00 in free credits. Always confirm current limits on AI21 Labs's official pricing page — free tiers change without notice.
$ How much does AI21 Labs cost per 1M tokens?
AI21 Labs does not publish simple per-1M-token pricing (it may bill per request, per output, or via a subscription plan). Check their official pricing page for exact figures.
$ Is AI21 Labs safe to use?
Uprouter rates AI21 Labs as low risk. Medium risk, driven by time-boxed economics rather than policy exposure. The free grant is $10 that expires after 7 days, so a stalled evaluation simply loses the balance, and no free tier is published for ongoing use. Cost variance between tiers is steep (Jamba Mini at $0.4 per 1M output tokens versus Jamba Large at $8), so an unmeasured model swap can multiply spend twentyfold. Rate limits and private-cloud terms are gated behind the Custom plan, which means capacity assumptions cannot be verified from public pages alone. Risk ratings are editorial, evidence-based, and never influenced by affiliate relationships.
Raw community sentiment, shown unblended. Votes never feed into objective data fields like price, risk, or status — and never into sort order.
No published reviews yet — be the first.
// more_informationprice history · live status · risk & confidence · code snippets
price_history
No recorded changes yet.
live_status
full history →best-effort probes · not a guarantee · last 15h ago
| Checked | Status | Latency |
|---|---|---|
| 15h ago | up | 84ms |
| 1d ago | up | 100ms |
| 3d ago | up | 101ms |
| 4d ago | up | 120ms |
| 6d ago | up | 87ms |
| 7d ago | up | 110ms |
| 9d ago | up | 139ms |
| 10d ago | up | 87ms |
| 12d ago | up | 78ms |
| 13d ago | up | 92ms |
| 14d ago | up | 73ms |
| 15d ago | up | 89ms |
| 15d ago | up | 101ms |
| 16d ago | up | 81ms |
| 17d ago | up | 101ms |
| 18d ago | up | 117ms |
| 18d ago | up | 106ms |
| 19d ago | up | 123ms |
| 20d ago | up | 90ms |
| 21d ago | up | 97ms |
| 21d ago | up | 84ms |
| 22d ago | up | 103ms |
| 23d ago | up | 78ms |
| 24d ago | up | 111ms |
| 25d ago | up | 132ms |
| 26d ago | up | 85ms |
| 26d ago | up | 94ms |
| 27d ago | up | 105ms |
| 28d ago | up | 99ms |
| 29d ago | up | 89ms |
Uprouter runs its own probes; best-effort, not a guarantee. Probe results depend on our network location and cadence — for production, always use the provider’s own status page.
risk_&_confidence
Medium risk, driven by time-boxed economics rather than policy exposure. The free grant is $10 that expires after 7 days, so a stalled evaluation simply loses the balance, and no free tier is published for ongoing use. Cost variance between tiers is steep (Jamba Mini at $0.4 per 1M output tokens versus Jamba Large at $8), so an unmeasured model swap can multiply spend twentyfold. Rate limits and private-cloud terms are gated behind the Custom plan, which means capacity assumptions cannot be verified from public pages alone.
Confidence reflects how much verifiable evidence (official docs, probe history, community reports) backs this entry. Risk is our editorial assessment of operator reliability and terms — not financial advice.
code_snippets
# AI21 Labs's own API — get a key from their dashboard
curl https://YOUR-PROVIDER-ENDPOINT/chat/completions \
-H "Authorization: Bearer THEIR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "jamba-large-1-7",
"messages": [{ "role": "user", "content": "Hello" }]
}'Direct to the provider — Uprouter receives nothing; verify their current pricing before purchasing.
Ranked by neutral similarity signals only — live status, free-tier shape and free-credit proximity. Commercial relationships never influence this list.