IBM Granite 4 Small-H (Ollama) is a hybrid Mamba and transformer model with a 1,000K context, 32B total parameters and roughly 9B active. Inside Atlas it is the model you pick when the window is the point: 1M tokens (1,048,576), Free (self-hosted), from a 19GB download. The hybrid state-space design keeps the memory curve far flatter than a pure attention model at the same length, which is what makes a million-token window locally reachable at all. Reserve roughly 21GB for the weights before the cache grows.
Why run IBM Granite 4 Small-H for a million-token context?
IBM Granite 4 Small-H (Ollama) offers 1M tokens (1,048,576) of context from a 19GB download, which no pure-transformer model in this size class comes close to offering locally. The million-token window is the whole reason to run it, and the hybrid Mamba architecture is what makes it affordable.
The -h suffix in granite4:32b-a9b-h marks the hybrid Mamba design. Long-context memory growth is far gentler than the quadratic KV cache of a standard attention stack, which is precisely why a 1M window is even discussable on consumer hardware. Inside Atlas that translates into sessions that do not compact. Atlas fans out work to subagents that can run in the foreground or in parallel background sessions, and long-running background subagents are exactly the workload that a 1M-token window protects. IBM Granite 4 Small-H is also 32B total with roughly 9B active per token, so it generates faster than a dense 32B while occupying a similar 21GB, which matters when an agent is producing many turns.
How much memory does IBM Granite 4 Small-H actually need?
IBM Granite 4 Small-H (Ollama) needs roughly 21GB of VRAM or unified memory for the weights alone, from a 19GB download. That 21GB figure is the weight floor, not the total: filling anything close to 1M tokens is still a serious memory commitment even with the hybrid design.
The honest reading of IBM Granite 4 Small-H is that the hybrid architecture changes the slope of the memory curve, not the fact that there is one. Raise num_ctx and watch memory as you approach long contexts rather than requesting the full 1,048,576 immediately. In practice most Atlas sessions never need the top of that range, and the value of the window is that you stop thinking about compaction for ordinary work rather than that you routinely fill it. Because IBM Granite 4 Small-H is Free (self-hosted), the experiment costs only time: raise the window, watch resident memory, and settle at the largest value your card sustains.
What can a 1M window do that a 128K window cannot in Atlas?
With IBM Granite 4 Small-H (Ollama) at 1M tokens (1,048,576), Atlas can put a very large slice of a private repository in context at once. Feed that window from a locally embedded index: with Ollama embeddings, nothing about the repo leaves the machine while you fill a million tokens.
Atlas can build its code index with local Ollama embeddings, keeping code off third-party servers, and Atlas indexes code by AST declarations using tree-sitter, not blind line windows. Combine those with IBM Granite 4 Small-H and the retrieval question changes shape: instead of agonizing over which five chunks fit, you can afford to be generous and let the model see the surrounding declarations, the tests, and the call sites together. Atlas searches code with hybrid semantic and keyword retrieval fused by reciprocal rank fusion, so the ordering still matters, but the penalty for a slightly imperfect ranking drops sharply when the window is 1M rather than 8K.
When should you pick a different model than IBM Granite 4 Small-H?
Skip IBM Granite 4 Small-H (Ollama) if you cannot spare roughly 21GB of VRAM or unified memory, or if your Atlas sessions are short. A 1M tokens (1,048,576) window that you never fill is 21GB of hardware spent on nothing, and a smaller long-context model will serve you better.
IBM Granite 4 Small-H is the right pick for whole-repository reasoning, very long tool transcripts, and background subagents that run for hours. It is the wrong pick for a laptop with 16GB of unified memory, or for quick single-file patches where a code-specialized model at a fraction of the size will produce a better diff. Atlas lets you switch the active model and provider on the fly with favorites and recents, so the sane pattern is to keep granite4:32b-a9b-h as the long-context favorite and switch away from it when a task is small. Everything here is Free (self-hosted), so the only real budget is memory.
Setup
- 01Pull the hybrid large tag: ollama pull granite4:32b-a9b-h (19GB, also tagged :small-h).
- 02Reserve roughly 21GB of VRAM or unified memory for the weights, then raise num_ctx and watch memory as you approach long contexts.
- 03In atlas.json register granite4:32b-a9b-h under the ollama provider with limit.context 1048576 and limit.output 16384.
- 04Set model to ollama/granite4:32b-a9b-h and confirm the 1M window resolved with atlas models ollama.
- 05Feed that window from a locally embedded index: with Ollama embeddings you can put a very large slice of a private repo in context and still have nothing leave the machine.
Frequently asked questions
- how to run granite 4 small-h in atlas
- Run ollama pull granite4:32b-a9b-h (19GB, also tagged :small-h), register it in atlas.json under the ollama provider with limit.context 1048576 and limit.output 16384, then set model to ollama/granite4:32b-a9b-h.
- what is granite 4 small-h's context window
- IBM Granite 4 Small-H (Ollama) has a 1M tokens (1,048,576) context. The hybrid Mamba and transformer design keeps the memory curve far flatter than a pure attention model at the same length.
- how much vram do i need for granite4 32b-a9b-h
- Reserve roughly 21GB of VRAM or unified memory for the weights of IBM Granite 4 Small-H. That is the floor: approaching 1M tokens (1,048,576) requires meaningfully more.
- what does the -h suffix mean in granite4
- The -h suffix marks the hybrid Mamba architecture. In IBM Granite 4 Small-H, long-context memory growth is far gentler than the quadratic KV cache of a standard attention stack.
- is granite 4 small-h faster than a dense 32b model
- Yes. IBM Granite 4 Small-H is 32B total with roughly 9B active per token, so it generates faster than a dense 32B while occupying a similar 21GB.
- how much does granite 4 small-h cost
- IBM Granite 4 Small-H (Ollama) is Free (self-hosted). The cost is hardware: a 19GB download and roughly 21GB of memory to serve the weights.
- why is the granite4 1b tag bigger than the 3b tag
- Ollama renders the granite4 1b tag at 3.3GB, larger than the 3b tag, which looks like a quantization mismatch rather than a real size. The Granite 4 line is new and the non-hybrid tags are less consistent.
Try Atlas in your terminal
The terminal-native AI coding agent. Free core, single binary.
Install AtlasRelated guides
Atlas vs CodeGPT in 2026: A Developer's Guide to Terminal and IDE AI Agents
Comparing Atlas and CodeGPT in 2026 for developers. Atlas offers terminal-native TUI and permission-gated tools, while CodeGPT provides IDE integration and a full repo Knowledge Graph.
Atlas for Flutter in 2026
Discover Atlas for Flutter in 2026. This terminal-native AI coding agent helps Flutter developers build apps faster and safer, integrating with widgets, state, and the Dart toolchain.
Atlas vs GitHub Copilot CLI: A Developer's Guide for 2026
Comparing Atlas and GitHub Copilot CLI for developers in 2026. Explore pricing, privacy, agentic workflows, and extensibility to choose the best terminal AI coding agent.
Atlas for Nim: A Terminal-Native AI Coding Agent for Nimble Packages and Macros in 2026
Atlas is a terminal-native AI coding agent for Nim in 2026. It reads .nimble requires and asterisk-exported symbols, adds std/unittest suites, runs nimble test, formats with nph.
Atlas for Gin in 2026
Atlas is a terminal-native AI coding agent for Gin in 2026. It reads router groups and binding tags, then runs go test ./... -race behind a permission prompt.
Atlas for Nuxt: Auto-Imports, useAsyncData, and Nitro Handlers in 2026
Atlas is a terminal-native AI coding agent for Nuxt in 2026. It reads nuxt.config.ts, pages/ routes, composables/ auto-imports, and server/api/ Nitro handlers, and tests with @nuxt/test-utils.
Atlas for Haskell in 2026
Atlas is a terminal-native AI coding agent for Haskell in 2026. Run it in a project with a .cabal file or a stack.yaml, let cabal build drive it, and review the diff.
Atlas for Fastify in 2026
Atlas is a terminal-native AI coding agent for Fastify in 2026. It reads plugin encapsulation and JSON schemas, then runs node --test behind a permission prompt.