Llama 3.3 70B Instruct
Offerings
| Host | Variant | Region | Status | Input $/Mtok | Output $/Mtok | Cache read | Cache write | Confidence? | Source |
|---|---|---|---|---|---|---|---|---|---|
| akashml | fp8 | – | active | $0.20 | $0.52 | $0.10 | – | api | openrouter |
| aws-bedrock | – | – | active | $0.72 | $0.72 | – | – | aggregator | litellm, modelsdev, backfill-litellm |
| aws-bedrock | – | us | active | $0.72 | $0.72 | – | – | aggregator | litellm |
| azure | – | – | active | $0.71 | $0.71 | – | – | aggregator | litellm, modelsdev, backfill-litellm |
| cloudflare | fp8 | – | active | $0.29 | $2.25 | – | – | api | openrouter |
| coreweave | – | – | active | $0.71 | $0.71 | $0.71 | – | api | openrouter |
| deepinfra | – | – | active | $0.23 | $0.40 | – | – | aggregator | litellm, modelsdev, backfill-litellm |
| deepinfra | fp8+turbo | – | active | $0.10 | $0.32 | – | – | api | openrouter |
| google-vertex | – | – | active | $0.72 | $0.72 | – | – | api | openrouter, litellm, modelsdev, backfill-litellm |
| google-vertex | – | us-central1 | active | $0.72 | $0.72 | – | – | api | openrouter |
| groq | – | – | active | $0.59 | $0.79 | $0.30 | – | api | openrouter, modelsdev |
| hyperbolic | – | – | active | $0.12 | $0.30 | – | – | aggregator | litellm, backfill-litellm |
| meta | – | – | active | free | free | – | – | aggregator | litellm, modelsdev, backfill-litellm |
| nebius | – | – | active | $0.13 | $0.40 | – | – | aggregator | litellm, backfill-litellm |
| novita | – | – | active | $0.14 | $0.40 | – | – | api | openrouter, litellm, modelsdev, backfill-litellm |
| nvidia | – | – | unlisted | free | free | – | – | aggregator | modelsdev |
| openrouter | – | – | active | $0.71 | $0.71 | $0.71 | – | api | openrouter |
| parasail | fp8 | – | active | $0.22 | $0.50 | $0.11 | – | api | openrouter |
| primeintellect | – | – | unlisted | $1.04 | $2.25 | – | – | api | openrouter |
| sambanova | – | – | active | $0.60 | $1.20 | – | – | api | openrouter, litellm, backfill-litellm |
| together | – | – | active | $1.04 | $1.04 | – | – | api | openrouter, litellm, modelsdev, backfill-litellm |
Confidence: official = vendor pricing page · api = OpenRouter per-host · aggregator = community list, may lag · unreviewed = automatic extraction awaiting review. When sources disagree the most authoritative wins. Details: /methodology#confidence
Sources for this model
| Source | URL | Last fetched | Confidence |
|---|---|---|---|
| openrouter | https://openrouter.ai/meta-llama/llama-3.3-70b-instruct | 2026-11-19 05:12 UTC | api |
| litellm | https://raw.githubusercontent.com/BerriAI/litellm/main/model_prices_and_context_window.json#novita/meta-llama/llama-3.3-70b-instruct | 2026-11-19 05:12 UTC | aggregator |
| modelsdev | https://models.dev/api.json#novita-ai/meta-llama/llama-3.3-70b-instruct | 2026-11-19 05:12 UTC | aggregator |
Cite this page
Llama 3.3 70B Instruct: $0.10/$0.32 per 1M tokens on deepinfra as of 2026-11-19. Token Barometer by Dejan Strbad, https://tokenbarometer.com, data CC BY 4.0. https://tokenbarometer.com/model/meta/llama-3.3-70b-instruct
Badge
Embed a pricing badge for this model:
[](https://tokenbarometer.com/model/meta/llama-3.3-70b-instruct)
Price history
Change log
| Date | Host | Kind | Change |
|---|---|---|---|
| 2026-11-14 | openrouter | Metadata changed | max output tokens 128000 → 115200 |
| 2026-11-13 | akashml | Metadata changed | max output tokens 115200 → 128000 |
| 2026-11-11 | openrouter | Metadata changed | max output tokens 128000 → 115200 |
| 2026-11-10 | akashml | Metadata changed | max output tokens 32768 → 128000 |
| 2026-11-10 | novita | Removed from source | Removed from source |
| 2026-11-07 | openrouter | Metadata changed | max output tokens 128000 → 32768 |
| 2026-11-05 | akashml | Metadata changed | max output tokens 16384 → 128000 |
| 2026-11-05 | together | Removed from source | Removed from source |
| 2026-11-04 | openrouter | Metadata changed | max output tokens 128000 → 16384 |
| 2026-11-03 | akashml | Metadata changed | max output tokens 16384 → 128000 |
| 2026-10-25 | meta | Removed from source | Removed from source |
| 2026-10-25 | openrouter | Metadata changed | max output tokens 128000 → 16384 |
| 2026-10-24 | nvidia | Removed from source | Removed from source |
| 2026-10-24 | akashml | Metadata changed | max output tokens 16384 → 128000 |
| 2026-10-18 | openrouter | Metadata changed | max output tokens 128000 → 16384 |
| 2026-10-17 | akashml | Metadata changed | max output tokens 32768 → 128000 |
| 2026-10-15 | openrouter | Metadata changed | max output tokens 128000 → 32768 |
| 2026-10-13 | akashml | Metadata changed | max output tokens 16384 → 128000 |
| 2026-10-12 | meta | Metadata changed | deprecation date 2026-10-21 → 2027-03-30 |
| 2026-10-10 | openrouter | Metadata changed | max output tokens 128000 → 16384 |
| 2026-10-08 | akashml | Metadata changed | max output tokens 16384 → 128000 |
| 2026-10-07 | openrouter | Metadata changed | max output tokens 128000 → 16384 |
| 2026-10-06 | primeintellect | Removed from source | Removed from source |
| 2026-10-06 | akashml | Metadata changed | max output tokens set to 128000 |
| 2026-09-24 | google-vertex | New offering | New offering |
| 2026-08-27 | deepinfra | Price cut | 0.13 → 0.10 (-23.1%), 0.39 → 0.32 (-17.9%) |
| 2026-08-25 | together | Price increase | 0.88 → 1.04 (+18.2%), 0.88 → 1.04 (+18.2%) |
| 2026-01-13 | novita | Price cut | 1080.00 → 0.14 (-100.0%), 3200.00 → 0.40 (-100.0%) |
| 2026-01-09 | novita | New offering | New offering |
| 2025-10-04 | deepinfra | Price increase | 0.0380 → 0.13 (+242.1%), 0.12 → 0.39 (+225.0%) |
| 2025-08-22 | deepinfra | New offering | New offering |
| 2025-07-21 | hyperbolic | New offering | New offering |
| 2025-06-06 | nebius | New offering | New offering |
| 2025-05-19 | sambanova | New offering | New offering |
| 2025-01-20 | together | New offering | New offering |
| 2024-12-21 | aws-bedrock | New offering | New offering |
| 2024-12-18 | azure | New offering | New offering |