> SambaNova
SambaNova Cloud serves open-weight models (DeepSeek-V3.1, DeepSeek-V3.2, Meta-Llama-3.3-70B-Instruct, gpt-oss-120b, gemma-4-31B-it, MiniMax-M2.7 and MiniMax-M3) through an OpenAI-compatible endpoint at api.sambanova.ai/v1, priced per 1M tokens from $0.22 input / $0.59 output for gpt-oss-120b up to $3 / $4.50 for DeepSeek-V3.1 and V3.2. The free tier is defined by the absence of a payment method: 20 requests per minute, 20 requests per day and 200,000 tokens per day per model, on the five models listed in the free-tier tables. Linking a card moves the account to the Developer tier, which raises per-model limits and caps the account at 20M tokens per day across all models.
// pros
- Free tier needs no payment method and applies automatically; published limits are 20 RPM, 20 RPD and 200,000 tokens per day for DeepSeek-V3.1, Meta-Llama-3.3-70B-Instruct and gpt-oss-120b, plus preview models DeepSeek-V3.2 and gemma-4-31B-it.
- Developer tier lifts per-model limits to 60 RPM / 12,000 RPD for DeepSeek-V3.1, gpt-oss-120b and MiniMax-M2.7 and 240 RPM / 48,000 RPD for Meta-Llama-3.3-70B-Instruct, with a 20M tokens/day account-wide cap.
- Cache reads are priced explicitly on MiniMax-M2.7 at $0.06 per 1M cached input tokens against $0.60 per 1M uncached, a 90% discount shown directly in the pricing table.
- Cheapest published rows are gpt-oss-120b at $0.22/1M input and $0.59/1M output and gemma-4-31B-it at $0.38/$1.15; every rate is metered per 1M tokens with no subscription fee.
- Each response carries rate-limit headers (x-ratelimit-limit-requests, -remaining-requests, -reset-requests and the per-day equivalents), so clients can back off before receiving a 429.
- The API is OpenAI-compatible (chat completions with streaming) against api.sambanova.ai/v1, so existing OpenAI SDK code runs after a base-URL and key change.
// cons
- Free tier is capped at 20 requests per day and 200,000 tokens per day per model, which is smoke-test volume; MiniMax-M2.7 and MiniMax-M3 have no free-tier rows at all.
- Preview models (DeepSeek-V3.2, gemma-4-31B-it) are documented as evaluation and experimentation only, with limited capacity and possible removal at short notice.
- Price spread inside one catalogue is wide: DeepSeek-V3.1 and V3.2 cost $3/1M input and $4.50/1M output, roughly 14x the input rate of gpt-oss-120b.
- Adding a payment method moves the account to Developer-tier pricing and limits, so a card added for verification changes both cost and quota at once.
- No published USD sign-up credit: the plans page only points to the developer community for opportunities to earn credits.
Use it when you want an OpenAI-compatible endpoint for Llama or gpt-oss class models at published per-token rates and can stay on the card-free tier for prototyping, then move to Developer once 20 requests a day is not enough. Pair it behind a router that selects by model price, because the DeepSeek rows cost several times the cheapest rows on the same table.
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: Free tier is a recurring daily allowance, not a USD credit: 200,000 tokens/day (20 RPM, 20 RPD) per the official rate-limits docs; 200k tokens at the cheapest published input rate of $0.22/1M (gpt-oss-120b) is about $0.04/day, while the same 200k at DeepSeek-V3.1's $3/1M input rate would be about $0.60/day.
| Plan | Type | Monthly | Input /1M | Output /1M | Note |
|---|---|---|---|---|---|
| Free tier (source-reported) | free | — | — | — | $5 free credits at signup (valid ~30 days to 3 months depending on the source) plus a persistent free tier; no credit card; rate limits ~10-30 RPM by model size. |
| Free Tier | free | — | — | — | Applied when no payment method is linked; includes rate-limited access. |
| DeepSeek-V3.1 PAYG | payg | — | $3.00 | $4.50 | Official rate for DeepSeek-V3.1 on SambaNova Cloud. |
| MiniMax-M2.7 PAYG | payg | — | $0.600 | $2.40 | Official rate for MiniMax-M2.7; input includes cached token rate at $0.06. |
Pricing normalized from public sources — always verify with the provider.
// model prices
| Model | Input /1M | Output /1M | Context |
|---|---|---|---|
| Llama 3.1 405B | — | — | — |
| DeepSeek-V3.1/V3.2 | — | — | — |
| gpt-oss-120b | — | — | 131k |
| Llama 3.3 70B Instruct | — | — | 131k |
| Qwen2.5 72B Instruct | — | — | 33k |
$ Does SambaNova have a free tier?
Yes — SambaNova offers a free tier that we value at roughly $0.04 in free credits. Always confirm current limits on SambaNova's official pricing page — free tiers change without notice.
$ How much does SambaNova cost per 1M tokens?
SambaNova 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 SambaNova safe to use?
Uprouter rates SambaNova as low risk. Medium. Free access is real but narrowly capped (20 RPD, 200,000 TPD) and is defined only by not linking a payment method, so cost exposure begins the moment a card is added and the Developer tier then caps the whole account at 20M tokens/day. SambaNova's privacy policy states personal data is transferred to and processed in the United States, with processors in the UK, EEA, Middle East and Asia Pacific, which EU teams should review before sending prompts. Preview models are explicitly removable at short notice. 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 | 301ms |
| 1d ago | up | 306ms |
| 3d ago | up | 329ms |
| 4d ago | up | 307ms |
| 6d ago | up | 301ms |
| 7d ago | up | 315ms |
| 9d ago | up | 493ms |
| 11d ago | up | 330ms |
| 12d ago | up | 341ms |
| 13d ago | up | 298ms |
| 14d ago | up | 300ms |
| 15d ago | up | 302ms |
| 15d ago | up | 326ms |
| 16d ago | up | 325ms |
| 17d ago | up | 337ms |
| 18d ago | up | 253ms |
| 19d ago | up | 297ms |
| 19d ago | up | 298ms |
| 20d ago | up | 300ms |
| 21d ago | up | 234ms |
| 22d ago | up | 293ms |
| 22d ago | up | 305ms |
| 23d ago | up | 315ms |
| 24d ago | up | 323ms |
| 25d ago | up | 312ms |
| 26d ago | up | 323ms |
| 27d ago | up | 313ms |
| 27d ago | up | 295ms |
| 28d ago | up | 322ms |
| 29d ago | up | 306ms |
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. Free access is real but narrowly capped (20 RPD, 200,000 TPD) and is defined only by not linking a payment method, so cost exposure begins the moment a card is added and the Developer tier then caps the whole account at 20M tokens/day. SambaNova's privacy policy states personal data is transferred to and processed in the United States, with processors in the UK, EEA, Middle East and Asia Pacific, which EU teams should review before sending prompts. Preview models are explicitly removable at short notice.
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
# SambaNova'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": "llama-3-1-405b",
"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.