Granite 3.1 1B A400m Instruct
Dec 2024
Granite 3.1 1B A400m InstructCompare inonMMLU-Pro · Official board
—per task?
self-host onlyprice
—output speed?
131Kcontext
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGemini 3.1 Prounstated×6.9—$2 / $12OpenGemma 2 9B It×3.9—$0.03 / $0.06Context: 131K → 8K+1
BetterGPT-4ounstated×5.6—$2.50 / $10Context: 131K → 128KBetterGPT OSS 20BmediumOpen×5.5—$0.018 / $0.09Memory (4-bit): 0.7 GB → 13 GB
BetterQwQ 32BOpen×5.2—$0.15 / $0.40Memory (4-bit): 0.7 GB → 20 GBModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
Gemini 3.1 Prounstated×6.9—$2 / $12GPT-4ounstated×5.6—$2.50 / $10Context: 131K → 128KGPT OSS 20BmediumOpen×5.5—$0.018 / $0.09Memory (4-bit): 0.7 GB → 13 GBQwQ 32BOpen×5.2—$0.15 / $0.40Memory (4-bit): 0.7 GB → 20 GB
Llama 3.3 70B InstructunstatedOpen×5—$0.10 / $0.32Context: 131K → 128K+1ModelScoreCostPer taskWhat changes
Gemma 2 9B ItOpen×3.9—$0.03 / $0.06Context: 131K → 8K+1
Phi 3.5 Mini InstructOpen×3.6——Memory (4-bit): 0.7 GB → 1.9 GBQwen 2 7B InstructOpen×3.6—$0.055 / $0.055Context: 131K → 33K+1LEXAONE 3.5 7.8B InstructOpen×3.5——Context: 131K → 33K+1
Mistral NemounstatedOpen×3.4—$0.15 / $0.15Memory (4-bit): 0.7 GB → 6.1 GBModelScoreCostPer taskWhat changes
Granite 4.2 8BthinkingOpenn/a—$0.024Context: 131K → 128K+1Granite 4.2 30BthinkingOpenn/a—$0.072Context: 131K → 128K+1Granite 4.2 3BthinkingOpenn/a—$0.006Context: 131K → 128K+1Granite 4.1 30BOpenn/a—$0.10 / $0.25Memory (4-bit): 0.7 GB → 17 GBGranite 4.1 3BOpenn/a——Memory (4-bit): 0.7 GB → 1.7 GBGranite 4.1 8Bnon-thinkingOpenn/a——Memory (4-bit): 0.7 GB → 4.4 GBGranite 4.0 350Mnon-thinkingOpenn/a——Context: 131K → 33KGranite 4.0 H 350MOpenn/a——Context: 131K → 33KAll IBM models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · full quality✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
Benchmarks
in Chat & KnowledgeMMLU-Pro · official board13.3#32
FAQ
How much does Granite 3.1 1B A400m Instruct cost?›
There is no per-unit API price: self-host only.
Is Granite 3.1 1B A400m Instruct free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Granite 3.1 1B A400m Instruct, and what is it best at?›
#32 of 32 in Chat & Knowledge.
Is Granite 3.1 1B A400m Instruct better than Mistral 7B Instruct v0.1?›
On MMLU-Pro · official board, Granite 3.1 1B A400m Instruct scores 13.3 vs 25.8 — lower.
What is the context window of Granite 3.1 1B A400m Instruct?›
131K tokens.
When was Granite 3.1 1B A400m Instruct released? Is there a newer version?›
Released 2024-12-06. It is the current version in its line.
Is Granite 3.1 1B A400m Instruct open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Granite 3.1 1B A400m Instruct locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Granite 3.1 1B A400m Instruct read images?›
No, it accepts text only.
What is Granite 3.1 1B A400m Instruct?›
Granite 3.1 1B A400m Instruct is an open-weights language model from IBM, released Dec 2024.
What are the best alternatives to Granite 3.1 1B A400m Instruct?›
Gemini 3.1 Pro (better), Gemma 2 9B It (open weights).