Jais Adapted 70B Chat
By MBZUAI / Core42 · UAE
Updated 2026-07-13
Overview
MBZUAI and Core42's Llama-2 70B extended with 32k Arabic tokens and GQA — the strongest open-weight Arabic LLM, reaching GPT-4-class quality in Arabic.
When to pick this model
- Production Arabic workloads needing top open quality
- Arabic legal, medical, or technical content generation
- Bilingual Arabic-English assistants at enterprise scale
- MENA sovereign deployments with multi-GPU budgets
- Replacing GPT-4 for Arabic-heavy regulated use cases
VRAM requirements by quantization
| Quantization | VRAM required |
|---|---|
| Q4_K_M (recommended) | 40 GB |
| Q5_K_M | 48 GB |
| Q8_0 | 75 GB |
| FP16 (no quantization) | 140 GB |
VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.
In practice, Jais Adapted 70B Chat spills past single consumer GPUs even at Q4_K_M (40 GB) — think dual-GPU or workstation cards. Stepping up to Q8_0 nearly doubles the footprint to 75 GB, and unquantized FP16 weights take 140 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.
Without a GPU, Jais Adapted 70B Chat needs roughly 64 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 1 tokens/sec on entry-level GPUs, on the order of 6 tokens/sec on a mid-range card, and up to 20 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.
What hardware do you need
The table below matches Jais Adapted 70B Chat to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.
| GPU memory | Example cards | Best fit for Jais Adapted 70B Chat |
|---|---|---|
| 8 GB | RTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GB | Does not fit — needs 40 GB at Q4_K_M |
| 12 GB | RTX 5070, RTX 5070 Ti Laptop, RTX 4080 Laptop | Does not fit — needs 40 GB at Q4_K_M |
| 16 GB | RTX 5080, RTX 4080 Super, Radeon RX 9070 XT | Does not fit — needs 40 GB at Q4_K_M |
| 24 GB | RTX 4090, Radeon RX 7900 XTX, RTX 5090 Laptop | Does not fit — needs 40 GB at Q4_K_M |
| 32 GB | RTX 5090 | Does not fit — needs 40 GB at Q4_K_M |
Which GPU should you buy to run Jais Adapted 70B Chat?
To run Jais Adapted 70B Chat locally at Q4, you need ~40 GB of VRAM. The best value for this is a Apple Mac Studio (64+ GB unified memory).
As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.
Strengths
- Strongest open-weight Arabic model available
- GPT-4-level performance in Arabic
- Jais license permits commercial use
- GQA improves inference efficiency at 70B scale
Limitations
- ~40 GB VRAM at Q4
- 4096-token context is restrictive for long documents
- Limited capability outside Arabic and English
Typical workloads
In our catalog grid, Jais Adapted 70B Chat is filed under Advanced Arabic, AR/EN Bilingual — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: multilingual workloads.
Note the 4k-token context window — fine for short interactions, limiting for long documents or big retrieval contexts. The Apache 2.0 license is permissive, so shipping it inside a commercial product raises no special legal questions.
Architecture & training
Architecture: Dense · 70B · specialized in Arabic + English · MBZUAI/Core42 UAE
Training: MBZUAI/Core42 — 395B token corpus of native Arabic + high-quality English.
The clear top pick for Arabic at 70B — choose it when GPT-4-grade Arabic must run on your own hardware.
Quick start
ollama pull hf.co/inceptionai/jais-instruct-GGUFOr use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.
Similar models worth comparing
Frequently asked questions
How much VRAM does Jais Adapted 70B Chat need?
At the recommended Q4_K_M quantization, Jais Adapted 70B Chat needs about 40 GB of VRAM. Q8_0 takes 75 GB, and unquantized FP16 weights take 140 GB.
Can Jais Adapted 70B Chat run without a GPU?
Yes — with roughly 64 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.
What context window does Jais Adapted 70B Chat support?
Jais Adapted 70B Chat supports a 4k-token context window (4,096 tokens).
Can I use Jais Adapted 70B Chat commercially?
Yes. Jais Adapted 70B Chat is released under Apache 2.0, a permissive open-source license that allows commercial use, modification and redistribution.
How fast is Jais Adapted 70B Chat on consumer hardware?
Our compatibility engine estimates on the order of 6 tokens/sec on a mid-range GPU and up to 20 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.
Which quantization of Jais Adapted 70B Chat should I download first?
Start with Q4_K_M (40 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. It does not fit a single 24 GB consumer card — plan for multi-GPU or server hardware.