SAVRN
Search Contact SAVRN

SAVRN Model Hub · Comparisons

gemma-4-E4B-it vs MiniCPM-o-4_5

Gemma-4-E4B-it has 8B parameters and MiniCPM-o-4_5 has 9.4B parameters; both are released under Apache License 2.0; at 16-bit, gemma-4-E4B-it needs about 19.2 GB (1x MI300X from $1.85 an hour) and MiniCPM-o-4_5 about 22.5 GB (1x MI300X from $1.85 an hour).

Published metadata for 2 models, each read from its own repository.
Field gemma-4-E4B-it
google/gemma-4-E4B-it
MiniCPM-o-4_5
openbmb/MiniCPM-o-4_5
Publisher Google OpenBMB
Task Any to any Any to any
Modality Multimodal Multimodal
Parameters, as reported 8B parameters 9.4B parameters
Architecture Gemma4ForConditionalGeneration MiniCPMO
Library transformers transformers
Context length 131,072 tokens 40,960 tokens
Repository size 16.0 GB 20.0 GB
Artifact formats safetensors safetensors, onnx
License apache-2.0 apache-2.0
Access Open weights, no gate Open weights, no gate
Memory at 16-bit (weights and margin) 19.2 GB 22.5 GB
Cheapest GPUs at 16-bit, per hour 1x MI300X, $1.85 1x MI300X, $1.85
Memory at 4-bit (weights and margin) 4.8 GB 5.6 GB
Cheapest GPUs at 4-bit, per hour 1x MI300X, $1.85 1x MI300X, $1.85
Revision viewed ee0ef6023621 503e754207c9
Downloads reported by the hub 4.5M 693.8k
Last observed 2026-09-18 2026-09-18

An evaluation row appears only where at least two of these models report the same benchmark with the same stated configuration, metric, unit and setup. Different evaluators stay named in each cell. Values are shown as reported: no unit conversion, no ranking.

Other Reported Results

These results are listed for each model on its own, because the conditions needed to compare them are not stated or do not match. Two results that leave a condition blank are not assumed to share it.

gemma-4-E4B-it

BenchmarkConditionsResultReported byRevisionDate
Idavidrein/gpqa Task diamondMetric diamondComparison conditions not established 58.6 Model Card
Reported by a third party
Evaluated revision not stated 2026-04-02
LiquidAI/ifstruct-v1.0 Task ifstruct_v1Metric ifstruct_v1Comparison conditions not established 76.65 Liquid AI — IFStruct v1.0 blog (gemma-4-E4B-it)
Reported by a third party
Evaluated revision not stated 2026-06-30
MMMU/MMMU_Pro Task mmmu_pro_visionMetric mmmu_pro_visionComparison conditions not established 52.6 Model Card
Reported by a third party
Evaluated revision not stated 2026-05-12
MathArena/aime_2026 Task MathArena/aime_2026Metric MathArena/aime_2026Comparison conditions not established 42.5 Model Card
Reported by a third party
Evaluated revision not stated 2026-03-02
TIGER-Lab/MMLU-Pro Task mmlu_proMetric mmlu_proComparison conditions not established 69.4 Model Card
Reported by a third party
Evaluated revision not stated 2026-04-02
llamaindex/ExtractBench Task longMetric longSetup Pipeline name: gemma4_e4b_vllm_extract_oneshot_structured_output_fileComparison conditions not established 26.82 ExtractBench
Reported by a third party
Evaluated revision not stated 2026-08-24
llamaindex/ExtractBench Task meanMetric meanSetup Pipeline name: gemma4_e4b_vllm_extract_oneshot_structured_output_fileComparison conditions not established 69.64 ExtractBench
Reported by a third party
Evaluated revision not stated 2026-08-24
llamaindex/ExtractBench Task mediumMetric mediumSetup Pipeline name: gemma4_e4b_vllm_extract_oneshot_structured_output_fileComparison conditions not established 48.93 ExtractBench
Reported by a third party
Evaluated revision not stated 2026-08-24
llamaindex/ExtractBench Task shortMetric shortSetup Pipeline name: gemma4_e4b_vllm_extract_oneshot_structured_output_fileComparison conditions not established 81.1 ExtractBench
Reported by a third party
Evaluated revision not stated 2026-08-24
llamaindex/ParseBench Task chartMetric chartSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 7.7 ParseBench
Reported by a third party
Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task layoutMetric layoutSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 35.8 ParseBench
Reported by a third party
Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task meanMetric meanSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 40.5 ParseBench
Reported by a third party
Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task tableMetric tableSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 25 ParseBench
Reported by a third party
Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task text_contentMetric text_contentSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 81.4 ParseBench
Reported by a third party
Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task text_formattingMetric text_formattingSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 52.5 ParseBench
Reported by a third party
Evaluated revision not stated 2026-04-14

SAVRN's Notes on gemma-4-E4B-it

The name says E4B, the file counts 8 billion parameters, and memory follows the second number: 16.0 GB of weights at 16-bit and 19.2 GB needed. That fits the cheapest setup on our Index, one 192 GB MI300X at $1.85 an hour on-demand, with room for nine more copies. You get an instruction-tuned model taking text, images and audio in and writing text out, so one deployment covers several channels.

On licensing, Apache 2.0 covers commercial use, modification and redistribution, provided the license notices and a statement of significant changes travel with it. Check the context: the publisher describes the family at up to 256K tokens, this checkpoint's configuration says 131,072, and 131,072 is what you deploy against. It derives from google/gemma-4-E4B; start there if you plan your own tuning. The scores come from the model card and outside evaluators, none from us, and no Index host prices it.

SAVRN's Notes on MiniCPM-o-4_5

We would look at this one when the job involves a camera and a microphone at the same time. OpenBMB assembled it from SigLip2 for vision, Whisper-medium for hearing, CosyVoice2 for speech and Qwen3-8B for language, and the feature they lead with is full-duplex multimodal live streaming. At 16-bit it needs 22.5 GB; 8-bit brings that to 11.2 GB and 4-bit to 5.6 GB. All three fit on one MI300X with 192 GB at $1.85 per hour, so the card decision is how many live sessions you want on it at once.

Apache License 2.0 applies: commercial use, modification and redistribution are permitted, and you keep the notices and state significant changes. Two checks. The context is 40,960 tokens, short for this hub, so plan session length around it. And the paper on file, arXiv:2408.01800, describes MiniCPM-V, so read the 4.5 card for what changed.

Questions

Which is larger, gemma-4-E4B-it or MiniCPM-o-4_5?

MiniCPM-o-4_5 (9.4B parameters) is larger than gemma-4-E4B-it (8B parameters), by the parameter counts their publishers report.

Which is cheaper to run, gemma-4-E4B-it or MiniCPM-o-4_5?

At 4-bit, gemma-4-E4B-it fits on 1x MI300X from $1.85 an hour and MiniCPM-o-4_5 on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use gemma-4-E4B-it commercially?

Yes. gemma-4-E4B-it is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Can I use MiniCPM-o-4_5 commercially?

Yes. MiniCPM-o-4_5 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Related Comparisons