Jamba 1.7 Large

AI21 Labs · ai21/ai21-jamba-large-1-7

Provider	AI21 Labs
Family	jamba
Type	llm-chat
Status	active
Released	2025-07-02
Updated	2026-02-02
Parameters	398.6B
Open weights	yes

What it fits on

Ordered by predicted speed. Bandwidth sets decode rate; memory decides whether it runs at all.

Every figure here is computed, not measured. Fit is weights at each quantisation against device memory, with a 25% allowance for the KV cache, activations and the OS. Decode rate is the memory-bandwidth roofline at 70% efficiency. Nobody has run this model on these devices.

Device	Memory	Bandwidth	Best quality	Weights	tok/s	Fastest
NVIDIA GB200 Grace Blackwell Superchip	372.0 GB	16000.0 GB/s	q5	249.1 GB	~45.0	~56.2 q4
AMD Instinct MI355X	288.0 GB	8000.0 GB/s	q4	199.28 GB	~28.1	~28.1 q4
Apple M3 Ultra	512.0 GB	819.0 GB/s	q6	298.92 GB	~1.9	~2.9 q4

Where it runs

Taken from the card's availability section.

Platform	Id on that platform
Hugging Face	ai21labs/AI21-Jamba-Large-1.7	gated

Verified benchmark evidence

Each row was checked against its source by a reviewer, and carries the model identifier as actually evaluated — which is not always the same as this card's.

Benchmark	Evaluated as	Score	Evidence date	Source kind
aa_lcr	Jamba 1.7 Large	19.33%	2026-09-10 evaluated	independent evaluator	source
critpt	Jamba 1.7 Large	0.0%	2026-09-10 evaluated	independent evaluator	source
gpqa_diamond	Jamba 1.7 Large	38.99%	2026-09-10 evaluated	independent evaluator	source

Reported benchmark scores

This card reports no benchmark scores yet.

Benchmark	Catalogue standing	Score

Data

This card as JSON · See it in the graph · Edit on GitHub