MDL
Llama 3 Instruct 70B
Meta
Verified against Artificial Analysis · Jul 13, 2026
Capability
35
percentile index
Price /M
$0.65 / $2.75
in / out
Context
—
Avg score
22.5
10 benchmarks
AA Intelligence
3.5
index
Speed
47 tok/s
TTFT 0.7s
Released
Apr 18, 2024
About
Llama 3 Instruct 70B is a open-weights model in the Llama family from Meta. Benchmarked on 10 evals, averaging 22.5.
Benchmark scores · 10
avg 22.5 — higher is betterThe research behind this model
all papers mentioning it →Agentic Abstention: Do Agents Know When to Stop Instead of Act?▲ 146 on HF · Jun 27, 2026Information-Aware KV Cache Compression for Long Reasoning▲ 11 on HF · Jun 25, 2026RoPE-Aware Bit Allocation for KV-Cache Quantization▲ 8 on HF · Jun 23, 2026Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning▲ 7 on HF · Jun 1, 2026KronQ: LLM Quantization via Kronecker-Factored Hessian▲ 6 on HF · Jul 8, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed47 tok/s
Latency (TTFT)0.7s
Price in / out$0.65 / $2.75 per 1M
Blended price$1.18 / 1M
Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →
History
full ledger →Capability over time
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$2.75/M