Can an RTX 5090 run Codestral-22B-v0.1?

The answer

llm-speed has not measured Codestral-22B-v0.1 on RTX 5090 itself yet, so there is no llm-speed verdict. 1 community-submitted run exists (fastest 93.7 tok/s at an unreported quant on llama.cpp); community-submitted, not measured by llm-speed, so it gives no verdict.

1 signed run on llm-speed. Single-request decode speed in tokens per second; the short-chat workload unless noted. This run is community-submitted (signed on the submitter's own machine, not measured by llm-speed, and not used for llm-speed verdicts). Verdicts and headline numbers use only llm-speed's own runs and fit tests.

Measured tok/s by quant and runtime

Quant	Runtime	Build	Decode tok/s (best run)	Workload	Runs	Source
not reported	llama.cpp	not reported	93.7	chat-short	1	Community-submitted

llm-speed has not measured this pair itself, so there is no llm-speed verdict or fastest quant. The rows above are community-submitted, not measured by llm-speed.

VRAM: not recorded for these runs. The VRAM calculator gives an estimate (predicted, not measured).

Every measured row

Run	Date (UTC)	Workload	Quant	Runtime / build	Prompt tokens	Output tokens	Decode tok/s	Prefill tok/s	TTFT ms	VRAM GiB	Source
r_4q040m4scic	2026-07-01	chat-short	–	llama.cpp · not reported	0	256	93.7	–	69	–	Community-submitted
r_4q040m4scic	2026-07-01	chat-long	–	llama.cpp · not reported	0	250	87.4	–	849	–	Community-submitted
r_4q040m4scic	2026-07-01	agent-trace	–	llama.cpp · not reported	3362	840	93.4	10351.0	191	–	Community-submitted
r_4q040m4scic	2026-07-01	concurrent-decode	–	llama.cpp · not reported	1024	256	100.3	–	–	–	Community-submitted

Questions