Skip to content
Local AI Wizard
Your machine
Theme

Granite 4.2 8B

IBM · ibm-granite/granite-4.2-8b

Q4 file 5.3 GB (Q4_K_M) · runs well from 12 GB · Weights license: Apache 2.0

LMArena · text1,290 ± 11Rank 294 · release 2026-10-22 · its own scale
Memory needed~6 GBQ4 weights + working room, before the machine's reserve
Parameters8 Bmanufacturer specification
Context131,072 tokensPublished for this model

Quality

LMArena · text · rating
1,290 ± 11Confidence interval 1,279–1,301 · 3,259 votes · rank 294 of the releaseLMArena 2026-10-22 · CC BY 4.0

No result in LiveBench 2026-06-25 Not in LiveBench 2026-06-25

Reported by the publisher

21 results · IBM's model card · read 2026-10-25
SWE Bench Multilingual
30.78
SWE Bench Pro
19.11
SWE Bench Verified
47.67
Terminal-Bench 2.1
20.56
τ³-bench
66.34
τ³-bench (AVG)
58.06
BFCL (v4)
52.39
ProfBench
41.20
Every reported result21
BirdBench
41.07
GDPval
1,189
AIME25
86.67
HMMT Feb25
78.33
GPQA
64.14
LiveCodeBench v6
73.24
SciCode
36.09
MMLU-Pro
74.04
MMLU-ProX lite (IBM)
61.06
Arena-Hard-V2
65.19
IFBench (prompt)
79.33
RULER 64K
80.99
RULER 128K
71.41
Across Granite 4.2 sizes21 shared results · same publisher and settings
Granite 4.2: results the publisher reports for each size under the same settings
Benchmark (publisher settings)3B8B30B
SWE Bench MultilingualThis size's card does not report this result under these settings30.7841.89
SWE Bench ProThis size's card does not report this result under these settings19.1133.29
SWE Bench VerifiedThis size's card does not report this result under these settings47.6757
Terminal-Bench 2.1This size's card does not report this result under these settings20.5629.24
τ³-bench50.9966.3468.05
τ³-bench (AVG)45.7858.0662.00
BFCL (v4)52.4152.3961.39
ProfBench32.1041.2042.90
BirdBenchThis size's card does not report this result under these settings41.0741.85
GDPvalThis size's card does not report this result under these settings1,1891,225
AIME2578.3386.6789.17
HMMT Feb2566.6778.3389.17
GPQA54.8064.1466.41
LiveCodeBench v669.7173.2475.77
SciCode24.1136.0938.76
MMLU-Pro67.8474.0477.60
MMLU-ProX lite (IBM)27.7861.0666.64
Arena-Hard-V234.9665.1967.93
IFBench (prompt)74.3379.3377.17
RULER 64K67.5280.9989.96
RULER 128K55.3071.4181.38

Values as printed on the card, with the publisher's own settings (card section: Evaluation Results); column “8B Dense”. Not comparable across publishers and not used in any ranking. Model card

Run it

Run it

File Q4_K_M · 5.3 GB · ibm-granite/granite-4.2-8b-GGUF · What is Q4?

ollama run hf.co/ibm-granite/granite-4.2-8b-GGUF:Q4_K_M
  1. Install Ollama
  2. Run the command above
  3. Ask: Explain what a context window is, in two sentences.

Runs on

Fast memory holds it 114

The planning estimate fits the memory the model can use quickly: graphics card VRAM, or unified memory within the macOS GPU share.

Show all 114

Within memory, outside the fast part 5

Within the machine's memory, outside its fast part: on a graphics card the layers that do not fit run on the CPU from PC RAM (the listed RAM, or 32 GB assumed); on a Mac, above the macOS default GPU share. Not a speed figure.

Show all 5

Granite 4.2 8B is a text model from IBM with 8B parameters. It was released on August 25, 2026 and reads up to 131,072 tokens of context. Its Q4 weights plus working room come to about 6 GB before the machine's reserve; the smallest reference machines whose fast memory holds that are Mac mini M2 · 8 GB unified (~8 GB with its reserve) and RTX 4060 · 8 GB VRAM (~8 GB with its reserve). We collected 18 downloadable GGUF and MLX files from 2 publisher sources; the most downloaded GGUF source is ibm-granite's. It has no result in LiveBench 2026-06-25. Its LMArena text rating is 1,290 ± 11 (release 2026-10-22), a separate human-preference scale.

Hardware profiles grouped by the Q4 planning estimate on each machine (dedicated GPUs reserve 2 GB; unified and system memory 25%, 2 to 8 GB): the fast memory holds it, it is within memory but outside the fast part (layers offloaded to PC RAM, or above the macOS GPU share), or the machine is one memory tier short. Each card opens its hardware profile. Not a measured fit or speed.

On the reference machines

Granite 4.2 8B on the reference machines
Where Q4 sits
Mac mini M2 · 8 GB unifiedFast memory holds itNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column~8 GBFast memory holds itNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column
Mac mini M4 · 16 GB unifiedFast memory holds itQ6_KGGUF · 7.2 GB · needs ~15 GB16,384 tokens~10 GBFast memory holds itQ6_KGGUF · 7.2 GB · needs ~15 GB16,384 tokens
Mac mini M4 · 24 GB unifiedFast memory holds itQ8_0GGUF · 9.3 GB · needs ~19 GB32,768 tokens~12 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~19 GB32,768 tokens
Mac mini M4 · 32 GB unifiedFast memory holds itQ8_0GGUF · 9.3 GB · needs ~21 GB65,536 tokens~14 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~21 GB65,536 tokens
Mac mini M5 Pro · 64 GB unifiedFast memory holds itQ8_0GGUF · 9.3 GB · needs ~21 GB131,072 tokens~14 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~21 GB131,072 tokens
Mac Studio M5 Max · 128 GB unifiedFast memory holds itQ8_0GGUF · 9.3 GB · needs ~21 GB131,072 tokens~14 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~21 GB131,072 tokens
Mac Studio · M5 Ultra · 36-core CPU / 80-core GPU · 512 GB unifiedFast memory holds itQ8_0GGUF · 9.3 GB · needs ~21 GB131,072 tokens~14 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~21 GB131,072 tokens
RTX 4060 · 8 GB VRAMFast memory holds itNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column~8 GBFast memory holds itNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column
RTX 3060 · 12 GB VRAMFast memory holds itQ5_K_MGGUF · 6.3 GB · needs ~11 GB8,192 tokens~8 GBFast memory holds itQ5_K_MGGUF · 6.3 GB · needs ~11 GB8,192 tokens
RTX 5080 · 16 GB VRAMFast memory holds itQ8_0GGUF · 9.3 GB · needs ~15 GB16,384 tokens~8 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~15 GB16,384 tokens
RTX 4090 · 24 GB VRAMFast memory holds itQ8_0GGUF · 9.3 GB · needs ~15 GB65,536 tokens~8 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~15 GB65,536 tokens
RTX 5090 · 32 GB VRAMFast memory holds itQ8_0GGUF · 9.3 GB · needs ~15 GB65,536 tokens~8 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~15 GB65,536 tokens
DGX H100 · 8 × 80 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~29 GB131,072 tokens~22 GBFast memory holds itQ8_0GGUF · 9.3 GB · needs ~29 GB131,072 tokens

The file column uses the site's quant ladder (Q8_0 to Q3_K_M, MLX 8 to 4-bit on a Mac): the largest collected file whose size × 1.2 plus the context cache and the machine's reserve fits the fast memory. Planning estimates, not measured runs.

Plan another quantization or context

Consider it for

Compare Granite 4.2 8B using its publisher card. No overall score appears in the LiveBench 2026-06-25 snapshot.

Check first

The Q4 weights plus working room come to about 6 GB from 8B stated parameters, before the machine's reserve (Runs on adds it per machine). It is not a measured minimum; context, batch size, runtime and quantization change actual memory use.

Versions to download

Files you can download for this checkpoint. The section opens on the most downloaded collected GGUF source, ibm-granite, at Q4_K_M. On a Mac it opens on the most downloaded MLX source. Sizes are the Hub's published bytes. What is Q4?

ibm-granite (official)

ibm-granite/granite-4.2-8b-GGUF · granite-4.2-8b-Q4_K_M.gguf · 5.3 GB

Run this file

ollama run hf.co/ibm-granite/granite-4.2-8b-GGUF:Q4_K_M
llama-server -hf ibm-granite/granite-4.2-8b-GGUF:Q4_K_M
lms get https://huggingface.co/ibm-granite/granite-4.2-8b-GGUF
  • mlx-lm: mlx-lm runs MLX repositories; pick the MLX version on a Mac

Quality scores on this page belong to the original checkpoint; these files are not scored or ranked. File sizes read 2026-10-23 · download counts read 2026-10-17 · commands: Hugging Face Hub docs: Use Ollama with any GGUF model · llama.cpp server README (-hf <user>/<model>[:quant]) · LM Studio docs: lms get with a Hugging Face URL · mlx-lm README (mlx_lm.generate --model)

Fine-tunes, adapters and other versions20 fine-tunes
Quantized versions
59
Fine-tunes
20
Adapters · merges
7 · 0

Browse them on the Hugging Face model tree · read 2026-10-23

Uncensored and abliterated versions1

Some community uploads remove the model's safety refusals. Quality may drop, the original publisher does not support them, claims about them are the uploader's, and the original license still applies. This site never recommends or ranks them.

1 derivative repositories have “uncensored” or “abliterated” in their name (read 2026-10-23).

Measured speed

No published measurement for this checkpoint

Tokens per second are shown only when a source publishes them with hardware, build and setup. They are never estimated.

How to run

The memory needed above is the Q4 load before a machine's reserve. Formula: parameters × 0.625 bytes, plus 20% working allowance; each machine then adds its system reserve: 25% of unified or system memory, 2 to 8 GB, or 2 GB per GPU on dedicated cards. This is neither measured VRAM nor checkpoint download size.

Q4 planning estimate only. Check checkpoint format, runtime support, KV cache, context, batch size and offload before choosing hardware.

Evidence level: Generic estimate only; no execution reproduced by this site.

Publisher sampling defaults
temperature 1 · top_p 0.95generation_config.json · read 2026-10-23source
Chat template markers
tool calls · thinkingmarkers found in the template, not a test of the feature · read 2026-10-23source
Before downloading
  1. Check the exact checkpoint and license linked here.
  2. Confirm a compatible runtime and precision or quantization for your platform.
  3. Allow for context, batch size and any extra media components beyond the weights.

Technical details

What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.

Checkpoint / version
ibm-granite/granite-4.2-8b
Task and modality
language and reasoning · text
Count and basis
8B · manufacturer specificationibm-granite/granite-4.2-8b model card (Hugging Face) · read 2026-10-17source
Context
131,072 tokens published for this modelibm-granite/granite-4.2-8b model card (Hugging Face) · read 2026-10-17source
Architecture
Denseibm-granite/granite-4.2-8b model card (Hugging Face) · read 2026-10-17source
Released
ibm-granite/granite-4.2-8b model card (Hugging Face) · read 2026-10-17source
Runtime and precision
No local execution verified.
Languages
en, de, es, fr, ja, pt, ar, cs, it, ko, nl, zhModel card metadata · read 2026-10-25source
Commercial use
Allowed Apache License 2.0 text (SPDX) · read 2026-10-18license text
Every stored field14 fields
Revision checked
f8de16cdcdbc6c779ca517604e050d82cc119e44
Architecture
GraniteForCausalLM
Library
transformers
Hub task tag
text-generation
Declared base models
ibm-granite/granite-4.1-8b-base
Formats
safetensors, pt
Access
Open download
Publisher kind
Official
Tensor evidence
floating-tensor-elements · 8.79B · BF16checked 2026-09-26
Repository created
2026-08-07creation is not the model's release
Repository updated
2026-09-04a commit is not a release date
Hub downloads (30 days) · likes
78,274 · 108this repository only · read 2026-10-25
Catalog state
Hub config or weights found; no inventory benchmark link

Sources

Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.

Catalog reviewed: 2026-10-17. Benchmark release: 2026-06-25; snapshot collected: 2026-09-26.