Llama 3.3 70B Instruct costs $0.900/1M input tokens and $0.900/1M output tokens as of 2026-06-27. Meta publishes this model with a 128K tokens context window. o10 lists Llama 3.3 70B Instruct on NVIDIA NIM and SambaNova at routing band C2. API slug meta-llama-3-3-70b-instruct. Verify list prices against the venue you call.
Last updated: 2026-06-27. Full catalog. Reference prices are dated snapshots. Compare total cost and measured task quality before choosing a route.
Llama 3.3 70B Instruct is a chat model from meta (slug: meta-llama-3-3-70b-instruct). Pick Llama 3.3 70B Instruct when you want Meta Llama on a fulfillment host (or your own GPUs) instead of a closed lab SKU. Published input pricing starts at $0.900/1M tokens. Context window: 128K tokens.
Catalog snapshot 2026-06-27.
Pick Llama 3.3 70B Instruct when you want Meta Llama on a fulfillment host (or your own GPUs) instead of a closed lab SKU. Llama 3.3 70B Instruct is a Llama instruct checkpoint. Smaller Llama SKUs cut cost; larger ones raise quality. Prove the swap in shadow.
Llama 3.3 70B Instruct is a Llama instruct checkpoint. Smaller Llama SKUs cut cost; larger ones raise quality. Prove the swap in shadow.
Llama 3.3 70B Instruct costs $0.900/1M input tokens and $0.900/1M output tokens in the gateway catalog snapshot (2026-06-27). Endpoint pricing may differ by fulfillment host.
Sourced from gateway catalog snapshot.
Llama 3.3 70B Instruct is configured on NVIDIA NIM and SambaNova. Known host IDs: nvidia_nim: nvidia_nim/meta/llama-3.3-70b-instruct; nvidia_nim: sambanova/Meta-Llama-3.3-70B-Instruct. This model is routable through o10 eval-gated routing.
Llama 3.3 70B Instruct supports a 128K tokens context window in the gateway catalog snapshot. Context limits may vary by fulfillment host endpoint.
Llama 3.3 70B Instruct is classified at o10 routing band C2. Bands group models by cost tier and eval profile: C0 (economy), C1 (standard), C2 (capable), C3/C4 (frontier). Production routing is enabled for this slug.
Llama 3.3 70B Instruct is attributed to source provider meta in the o10 catalog. Fulfillment hosts (NVIDIA NIM and SambaNova) supply API access; list pricing may differ by venue.
Llama 3.3 70B Instruct in the o10 unified catalog.
| Field | Value |
|---|---|
| o10 slug | `meta-llama-3-3-70b-instruct` |
| Type | text |
| Band | C2 |
| Routable | Yes |
| Source provider | meta |
| Primary host | NVIDIA NIM |
| Context window | 128,000 tokens |
Gateway snapshot pricing for Llama 3.3 70B Instruct.
| Unit | Price |
|---|---|
| Input | $0.9/1M tokens |
| Output | $0.9/1M tokens |
Upstream venues configured for Llama 3.3 70B Instruct.
| Host | Upstream model ID |
|---|---|
| nvidia_nim | nvidia_nim/meta/llama-3.3-70b-instruct |
| nvidia_nim | sambanova/Meta-Llama-3.3-70B-Instruct |
Eval-gated model selection for this slug.
o10 slug: meta-llama-3-3-70b-instruct. Band: C2. Routable: yes.
See [/models/routing](/models/routing) for smart aliases (o10/auto, o10/frontier, o10/squad) and [/kyi](/kyi) for governance.
Fulfillment catalog 2026-06-27. Gateway pricing joined where slug match exists.
Llama 3.3 70B Instruct costs $0.900/1M input tokens and $0.900/1M output tokens in the gateway snapshot (2026-06-27). Verify against your provider's published pricing.
Llama 3.3 70B Instruct supports 128K tokens in the gateway catalog snapshot. Host-specific endpoints may differ.
Llama 3.3 70B Instruct (meta-llama-3-3-70b-instruct) is configured on: NVIDIA NIM and SambaNova. Upstream model IDs are listed in the fulfillment routes section.
Yes. Llama 3.3 70B Instruct is routable at band C2. Send model: "meta-llama-3-3-70b-instruct" or use an o10 smart routing alias.
Use o10 slug "meta-llama-3-3-70b-instruct" in the model field. See /models/routing for smart aliases (o10/auto, o10/frontier, o10/squad).
o10 routes Llama 3.3 70B Instruct only when per-use-case evals clear at your quality floor. Band C2 indicates cost tier. See /kyi for governance and /models/routing for smart aliases.
The o10 slug is "meta-llama-3-3-70b-instruct". Host IDs: nvidia_nim: nvidia_nim/meta/llama-3.3-70b-instruct; nvidia_nim: sambanova/Meta-Llama-3.3-70B-Instruct. Pricing snapshot date: 2026-06-27.
Llama 3.3 70B Instruct is hosted as nvidia_nim: nvidia_nim/meta/llama-3.3-70b-instruct; nvidia_nim: sambanova/Meta-Llama-3.3-70B-Instruct. o10 maps those upstream IDs to slug "meta-llama-3-3-70b-instruct".