Model

Pharia-1-LLM-7B (control)

OPEN

Aleph Alpha · Pharia-1 family · released Aug 26, 2024

Open, EU-compliance-focused 7B trained on lawfully sourced data; Aleph Alpha's strategic successor to Luminous.

ReasoningCodingVisionFunction callingTool useAgentic

-97.7

Elo · rank #412

Parameters

7.04B

Active params

7.04B (dense)

Context

8K tokens

Architecture

Dense decoder-only transformer, 27 layers, 4-head grouped-query attention

License

Open Aleph License (non-commercial research/educational use)

Languages

API price (in/out)

No hosted API

Modalities

text

Benchmark results

Bar shows position within the tracked field; marker = field best

AlpacaEvalHuman preference33.2%#31

best: Tülu 2+DPO 70B · 95.1%

ARC-ChallengeReasoning54.6%#110

best: Llama 3.1 405B · 96.9%

GSM8KMath1.4%#195

best: Llama 3.1 405B · 96.8%

HellaSwagReasoning64.6%#135

best: Claude 3 Opus · 95.4%

MMLU (EU-21 languages)Knowledge35.3%#16

best: Llama 3.1 70B · 77.1%

MMLUKnowledge48.4%#236

best: OpenAI o3 · 92.9%

TruthfulQAKnowledge54.7%#39

best: Phi-3.5-MoE (16x3.8B, 6.6B active) · 77.5%

WinoGrandeReasoning65.1%#123

best: PaLM 2 · 90.9%

Run it locally

VRAM @ Q4

4.5 GB

VRAM @ FP16

14.5 GB

Fits on (Q4)

RTX 3060 12GBRTX 4070 Ti 16GBRTX 3090 24GBRTX 4090 24GBRTX 5090 32GBM4 Pro 48GBM3 Max 128GBM3 Ultra 512GBA100 80GBH100 80GBH200 141GBB200 192GB

Throughput data unavailable.

Quantizations

—

Pharia-1 family

Elo progression across releases

Pharia-1-LLM-7B (control)Aug 2024-97.7

Resources

HFAleph-Alpha/Pharia-1-LLM-7B-control DOCSdocs.aleph-alpha.com/products/pharia-1-llm/overview

API price weights · each benchmark row carries its own source badge (see methodology)