SSyncDev

Jais-3-MoVA-36B-A4B

SyncDev model directory · Profile updated Nov 17, 2026

Compare with another model

Jais-3-MoVA-36B-A4B is the sparse Jais 3 model: 36B total, 4B active per token, 512K context, tuned for Arabic-English agents and tool use on one H100 node.

Key specifications

Specification score
53.4
Context window
524.29K
License
Apache-2.0

The specification score summarizes recorded features and limits. It is not a benchmark result.

Inputs and outputs

Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.

Accepts

  • Text input

Produces

  • Text output

Capabilities

ReasoningFunction callingOpen weights

What is Jais-3-MoVA-36B-A4B?

Thirty-six billion parameters stored, four billion doing the work on any given token. Jais-3-MoVA-36B-A4B is the sparse branch of IFM's Jais 3 family, and the reason to look at it is simple: roughly the quality of the dense 32B sibling at a fraction of the compute per token.

MBZUAI's Institute of Foundation Models and G42's Inception42 released it under Apache-2.0.

What "MoVA" means, as far as anyone has said

IFM calls the attention design Mixture-of-Values attention, MoVA, and stops there. It does not describe the mechanism or give an expert count. We will not guess. What you can see is its effect on serving: the validated recipe needs expert parallelism.

Sparse versus dense, on the same box

Both this model and the dense 32B fit one H100 node with two GPUs. The difference is what each token costs.

MoVA-36B-A4B Dense 32B
Active parameters per token 4B 32B
Stored parameters 36B (metadata: 37B) 32B (metadata: 35B)
Parallelism TP=2 plus EP=2 TP=2
Arabic IFBench (strict) 92.0 88.7
MCPMark 69.3 66.1
BrowseComp-Plus 71.8 71.9
Terminal Bench 2.1 67.3 63.7
Arabic MMLU 87.7 88.6
AraGen overall 68.6 71.0

Publisher-reported numbers. The sparse model edges ahead on tool and agent benchmarks and slips slightly on knowledge and open-ended Arabic generation. Against the wider field, [Qwen3.8-27B] remains stronger on coding agents in IFM's own comparison, scoring 79.8 on Terminal Bench.

Running it

  • vLLM 0.30.0+: --tensor-parallel-size 2 --enable-expert-parallel, plus the k2_horizon reasoning and tool-call parsers.
  • SGLang 0.5.20+: --tp 2 --ep 2, FlashAttention-3, and a model-override argument setting xllm_source_router_gemm_partitions to 2.
  • Always send reasoning_effort: "high"; the recommended output budget is at least 32,768 tokens.
  • FP8, GGUF and MLX builds are published under the IFM organisation.

Expert parallelism adds an operational wrinkle the dense model does not have. Check that your serving stack's version supports it before you schedule a migration.

Best fit

High-throughput Arabic-English agents: support desks that call several tools per conversation, browsing assistants, MCP-heavy workflows. Its best scores are precisely there. For legal drafting or knowledge-heavy Q&A the dense sibling is marginally safer.

The family's launch is reported for 15 October 2026; these weights had been public since 29 September.

Making the call

Run the same 200 production prompts through this model and the dense 32B on identical hardware. Record tokens per second at your target concurrency, cost per completed task, and pass rate on your tool calls. If the pass rates are within a point, the sparse model's throughput usually decides it. For broader comparisons try Gemma 4 26B A4B, another 4B-active model, or the compare tool.

How to evaluate Jais-3-MoVA-36B-A4B

Compare the model on the work you need to complete.

Tasks and inputs

Jais-3-MoVA-36B-A4B is listed for Text generation, Translation. It accepts text and produces text. Use the publisher’s input format and preprocessing for your chosen deployment.

Capacity and cost

Documented limits: 524.29K context. Measure a completed task using the prompts, documents and tools your application will use. Include reasoning and retries in the result.

Choosing a deployment

The weights are downloadable under Apache-2.0. Check runtime support and size the complete model and working context for your expected concurrency.

Frequently asked questions

About 4B of its 36B parameters per token.

These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.

Tencent

Hy4 Flash

Tencent Hy4 Flash is the cost-focused companion to Hy4, keeping vision and agent skills at a far lower token price. Availability is still unsettled.

1.05M context

Alibaba

Qwen4 Max

Qwen4 Max (qwen4-max) is Alibaba's first Qwen4-generation flagship: 1M-token context, text, image and video input, thinking on by default, $1.50/$4.50 per 1M.

1M context

Sarvam

Sarvam-105B-1711

Sarvam-105B-1711 is a refreshed open post-train of Sarvam's 105B MoE, tuned for tool calling, instruction following, agents and Indian languages.

— context

IBM

Granite 4.1 8B LLMON Preview

Granite 4.1 8B LLMON Preview is IBM's model trained on block-structured LLMON prompts to resist indirect prompt injection in agents and RAG.

— context

NVIDIA

Nemotron-Labs Sigma 8B

Nemotron-Labs Sigma 8B is NVIDIA's open continuous diffusion language model, denoising 64-token blocks in parallel; English, GPU-only, 8K context.

8.19K context

Alibaba

Qwen4 Max Preview

Qwen4 Max Preview was Alibaba's five-day early-access build of its Qwen4 flagship, offered only through a Token Plan subscription before GA qwen4-max.

1M context