Llama 3.3 70B
Available meta/llama-3.3-70b · by Meta · decoder-only-transformer
Pricing — 2 offering(s)
Groq
Input tokens
- $0.59 / 1M tokens (input) Current 2026-06-12 → present
Output tokens
- $0.79 / 1M tokens (output) Current 2026-06-12 → present
Together AI
Input tokens
- $1.04 / 1M tokens (input) Current 2026-06-12 → present
Output tokens
- $1.04 / 1M tokens (output) Current 2026-06-12 → present
Showing the active price and any recorded history. Full pricing history is available via the paid API — see API docs.
Capability profile
reasoning moderate
coding moderate
tool use moderate
instruction following strong
context window strong
multilingual moderate
speed strong
Operator guidance
The leading open-weight 70B option — route here for control, portability, and low hosted cost. Step up to a frontier closed model for the hardest tasks.
Use cases
- Open-weight self-hosting or low-cost hosted inference (Groq, Together AI)
- Workloads needing data control or no per-token vendor lock-in
Limitations
- Below current frontier closed models on the hardest reasoning/coding
- Quality/price depends on the serving provider, not Meta
- Capability ratings are qualitative, not a single cited benchmark run