Granite 4.2 30B
Granite 4.2 30BAlternatives?
CheaperLing 3.0 Flash VLthinkingOpen+16%-68%$0.06 / $0.18Memory (4-bit): 18 GB → 75 GB
BetterMiniMax M3thinkingOpen+39%+608%$0.51Memory (4-bit): 18 GB → 256 GB+2OOpenMiniCPM5 1Bnon-thinking×28.4——
CheaperMiMo v2.5 ProthinkingOpensame-25%$0.054Memory (4-bit): 18 GB → 614 GBCheaperGranite 4.2 3BthinkingOpensame-92%$0.006Ling 3.0 Flash VLthinkingOpen+16%-68%$0.06 / $0.18Memory (4-bit): 18 GB → 75 GBMiMo v2.5 ProthinkingOpensame-25%$0.054Memory (4-bit): 18 GB → 614 GBGranite 4.2 3BthinkingOpensame-92%$0.006
Gemma 4 E4BthinkingOpen-17%-89%$0.008MiMo v2.5thinkingOpen-20%-73%$0.019Memory (4-bit): 18 GB → 186 GBMiniMax M3thinkingOpen+39%+608%$0.51Memory (4-bit): 18 GB → 256 GB+2Gemini 4 Argonhigh+70%+2,677%*$1.99Synchronous output stops at 65,536+2MQuasar 438B+19%+2,725%$2.02
Qwen3.8 27Bnon-thinkingOpen+41%+3,371%$2.49max_tokens range · QwenCloud+4OMiniCPM5 1Bnon-thinkingOpen×28.4——Qwen3.8 27Bnon-thinkingOpen+41%+3,371%$2.49max_tokens range · QwenCloud+4Granite 4.0 Micronon-thinkingOpen-73%—$0.017 / $0.11Granite 4.1 3BOpen-73%——Granite 4.1 30BOpen-73%—$0.10 / $0.25Granite 4.0 1Bnon-thinkingOpen-73%——Granite 4.0 H 350MOpen-71%——Context: 128K → 33KGranite 4.0 H Smallnon-thinkingOpen-71%—$0.05 / $0.22Memory (4-bit): 18 GB → 19 GBGranite 4.1 8Bnon-thinkingOpen-71%——Granite 4.0 H 1BOpen-69%——All IBM models →Run it yourself
estimateHistory
2 versionsBenchmarks
in Chat & KnowledgeFAQ
How much does Granite 4.2 30B cost?›
$0.16 per 1M input tokens and $0.65 per 1M output tokens (cheapest host: DeepInfra). A typical benchmark task costs about $0.072 at default effort.
Is Granite 4.2 30B free to use?›
There is no free API; the cheapest paid price is $0.16 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Granite 4.2 30B, and what is it best at?›
#169 of 334 in Chat & Knowledge, #75 of 132 in Coding, #103 of 133 in Agents & Tool Use, #150 of 195 in Writing & Language, #107 of 133 in Vision & Documents.
Is Granite 4.2 30B better than Granite 4.1 30B?›
On AA Omniscience · Hallucination Rate, Granite 4.2 30B scores 25.6 vs 95 — higher.
How can I use Granite 4.2 30B?›
Through the API of DeepInfra.
What is the context window of Granite 4.2 30B?›
128K tokens.
When was Granite 4.2 30B released? Is there a newer version?›
Released 2026-08-25. It is the current version in its line.
Is Granite 4.2 30B open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Granite 4.2 30B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Granite 4.2 30B read images?›
No, it accepts text only.
What is Granite 4.2 30B?›
Granite 4.2 30B is an open-weights language model from IBM, released Aug 2026, the successor to Granite 4.1 30B.
What are the best alternatives to Granite 4.2 30B?›
Ling 3.0 Flash VL (cheaper), MiniMax M3 (better), MiniCPM5 1B (open weights).