OpenAI compatible API. Attested gateway. Public status.

Meta: Llama 3.3 70B Instruct Benchmarks

Benchmark and measurement links for Meta: Llama 3.3 70B Instruct, with TrustedRouter route data first.

Verify gateway
1 URLbase_url migration
100smodels and routes
0prompt logs by default

meta-llama/llama-3.3-70b-instruct

open weights Benchmarks

All models

Published benchmark scores

Benchmark scores for Meta: Llama 3.3 70B Instruct — every row links to its source, and a score is only ever attached to the exact checkpoint it was measured on. Vendor model-card and open-leaderboard numbers are cited, not run by us. Rows marked TrustedRouter · replays published are our own runs of this model through the gateway, with the full per-item replay published in trustedrouter-benchmarks so anyone can re-grade them.

BenchmarkCategoryScoreSource
HumanEval Coding 88.4% Meta — Llama 3.3 70B model card
2024-12-06
IFEval Instruction following 92.1% Meta — Llama 3.3 70B model card
2024-12-06
MMLU
0-shot, CoT
Knowledge 86.0% Meta — Llama 3.3 70B model card
2024-12-06
MMLU-Pro
CoT
Knowledge 68.9% Meta — Llama 3.3 70B model card
2024-12-06
MATH
0-shot, CoT
Math 77.0% Meta — Llama 3.3 70B model card
2024-12-06
GPQA Diamond
0-shot, CoT
Science 50.5% Meta — Llama 3.3 70B model card
2024-12-06

TrustedRouter measurements

TrustedRouter publishes route and status measurements without storing prompt or output content. Provider latency and uptime are exposed through the model performance and uptime pages.

External benchmark references

Sign in

Choose a sign in method.