IBM

Granite 4.2

IBM

Granite 4.2 adds native reasoning and reasoning-guided tool use to IBM’s open enterprise language models. The release includes 3B, 8B, and 30B checkpoints with full, low-effort, and non-thinking modes, allowing developers to select a latency and reasoning tradeoff. Multi-stage reinforcement learning targets coding, mathematics, science, and agent execution. All sizes natively support 128K context; the 30B checkpoint additionally supports a documented extension to 512K. Apache-licensed weights enable private deployment and customization. The family page groups sizes while preserving their individual checkpoint names as variants.

Granite SWASH

IBM

Granite SWASH is IBM Research’s open architectural preview for efficient language modeling. The family explores sliding-window attention at every layer, learnable attention sinks, and hybrid transformer designs, with dense 2B and sparse 3B checkpoints. The sparse model activates approximately 600M parameters. These are general-purpose English base models for research and customization, rather than instruction-ready assistants. Apache-licensed weights allow study of the architecture and downstream training; chat and tool behavior should not be inferred from the base release.

Granite Switch 4.1 Preview

IBM

Granite Switch 4.1 Preview adds explicit switching between ordinary language behavior and specialized programmatic capabilities. It is designed to combine a general model with deterministic functions through IBM’s Granite Libraries and Mellea tooling. The release includes 3B, 8B, and 30B language checkpoints under Apache 2.0, with 128K context for the representative 30B model. This is a publicly released preview, distinct from the ordinary Granite 4.1 family. Applications must supply the associated runtime to use specialized functions reliably.

Granite 4.1

IBM

Granite 4.1 is IBM’s enterprise language model release focused on instruction following and tool calling without requiring long reasoning traces. The language family offers 3B, 8B, and 30B checkpoints, trained with staged data refinement and released under Apache 2.0. This page represents the language checkpoints; vision, speech, embedding, and guardrail products in the same collection have different scopes. The models support local and managed deployment, allowing organizations to build private assistants and agent workflows around documented model cards rather than relying on an opaque hosted-only service.

Granite Vision 4.1

IBM

Granite Vision 4.1 is IBM’s 4B vision-language model for technical document understanding. It generates language answers from images, charts, tables, and diagrams, combining a compact language backbone with a vision encoder. The release targets enterprise workflows where users need grounded extraction and interpretation, rather than image generation. Apache-licensed weights support private deployment. Its model card documents task-specific evaluation and deployment instructions; results on chart benchmarks should not be treated as guaranteed accuracy on arbitrary business documents.

Granite 4.0

IBM

Granite 4.0 is IBM’s hybrid enterprise language family, combining Mamba-style state-space layers and attention to reduce long-context memory. Its sparse Small and Tiny tiers target retrieval and agents; dense Micro and later Nano checkpoints serve lower-resource applications. The initial release included Apache-licensed weights and watsonx access. This family page groups size tiers while documenting that Nano was released later in October, rather than assigning all future checkpoints the first announcement date.

Granite 3.3

IBM

Granite 3.3 is IBM’s open-weight language model release for enterprise applications. This reasoning and instruction update improves mathematical, coding, and conversational tasks, retaining explicit thought/response tags and a 128K context. It is distinct from speech-only or guardian classifiers. Apache-licensed weights support local inference and customization. Official model cards identify the checkpoint scope and release date. Enterprise uses still require application-specific evaluation, retrieval for current facts, and validation of generated code or structured outputs. Availability of downloadable weights is recorded separately from changing hosted API product lifecycles.

Granite 3.2

IBM

Granite 3.2 is IBM’s open-weight language model release for enterprise applications. This reasoning-focused update adds controllable thinking to the language generation, strengthening mathematics, coding, and instruction tasks. Reasoning traces are enabled through model-specific prompt settings. Apache-licensed weights support local inference and customization. Official model cards identify the checkpoint scope and release date. Enterprise uses still require application-specific evaluation, retrieval for current facts, and validation of generated code or structured outputs. Availability of downloadable weights is recorded separately from changing hosted API product lifecycles.

Granite 3.1

IBM

Granite 3.1 is IBM’s open-weight language model release for enterprise applications. This materially updated language generation improves long-context understanding, tool use, and multilingual instruction following, with 128K context. It groups ordinary base/instruct and size variants. Apache-licensed weights support local inference and customization. Official model cards identify the checkpoint scope and release date. Enterprise uses still require application-specific evaluation, retrieval for current facts, and validation of generated code or structured outputs. Availability of downloadable weights is recorded separately from changing hosted API product lifecycles.

Granite 3.0

IBM

Granite 3.0 is IBM’s open-weight language model release for enterprise applications. The third-generation language family adds stronger instruction following, retrieval-augmented generation, code understanding, and compact sparse models. The cited launch 8B Instruct checkpoint supports 4,096 tokens; the 128K extension belongs to the later Granite 3.1 release. Apache-licensed weights support local inference and customization. Official model cards identify the checkpoint scope and release date. Enterprise uses still require application-specific evaluation, retrieval for current facts, and validation of generated code or structured outputs. Availability of downloadable weights is recorded separately from changing hosted API product lifecycles.

Granite Code

IBM

Granite Code is IBM’s open-weight language model release for enterprise applications. The code foundation family provides base and instruction models for generation, explanation, repair, and translation across programming languages. The source cards explicitly date the initial release to May 6. Apache-licensed weights support local inference and customization. Official model cards identify the checkpoint scope and release date. Enterprise uses still require application-specific evaluation, retrieval for current facts, and validation of generated code or structured outputs. Availability of downloadable weights is recorded separately from changing hosted API product lifecycles.