Model
Model explorer
Jais-13B
OPENInception (G42) · Jais family · released Aug 30, 2023
First flagship open Arabic-English bilingual LLM, built with MBZUAI and trained on Cerebras' Condor Galaxy supercomputer.
ReasoningCodingVisionFunction callingTool useAgentic
-409.7
Elo · rank #443
Parameters
13B
Active params
13B (dense)
Context
2K tokens
Architecture
Dense GPT-3-style decoder-only transformer, trained from scratch with SwiGLU activations and ALiBi position encoding
License
Apache 2.0
Languages
2+
API price (in/out)
No hosted API
Modalities
text
Benchmark results
Bar shows position within the tracked field; marker = field best
best: Llama 3.1 405B · 96.9%
best: Claude 3 Opus · 95.4%
best: OpenAI o3 · 92.9%
best: GPT-4o mini · 93.1%
best: Phi-3.5-MoE (16x3.8B, 6.6B active) · 77.5%
Run it locally
VRAM @ Q4
9 GB
VRAM @ FP16
27 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
GGUF Q4 · GPTQ
Resources
API price weights · each benchmark row carries its own source badge (see methodology)