SAVRN Model Hub · Comparisons
medgemma-4b-it vs Qwen3-VL-4B-Instruct
Medgemma-4b-it has 4.3B parameters and Qwen3-VL-4B-Instruct has 4.4B parameters; medgemma-4b-it is released under other and Qwen3-VL-4B-Instruct under Apache License 2.0; at 16-bit, medgemma-4b-it needs about 10.3 GB (1x MI300X from $1.85 an hour) and Qwen3-VL-4B-Instruct about 10.7 GB (1x MI300X from $1.85 an hour).
| Field | medgemma-4b-it google/medgemma-4b-it | Qwen3-VL-4B-Instruct Qwen/Qwen3-VL-4B-Instruct |
|---|---|---|
| Publisher | Qwen | |
| Task | Image and text to text | Image and text to text |
| Modality | Image and text | Image and text |
| Parameters, as reported | 4.3B parameters | 4.4B parameters |
| Architecture | Gemma3ForConditionalGeneration | Qwen3VLForConditionalGeneration |
| Library | transformers | transformers |
| Context length | Not stated | 262,144 tokens |
| Repository size | 8.6 GB | 8.9 GB |
| Artifact formats | safetensors | safetensors |
| License | other | apache-2.0 |
| Access | Access requested at publisher | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 10.3 GB | 10.7 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 2.6 GB | 2.7 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 290cda5eeccb | ebb281ec70b0 |
| Downloads reported by the hub | 1.1M | 3.8M |
| 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.
Qwen3-VL-4B-Instruct
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| tiiuae/PBench | Task averageMetric averageSetup Detection-only model combined with SAM2 to convert boxes to segmentation masks.Comparison conditions not established | 49.2 | Community Evals Reported by a third party |
Evaluated revision not stated | 2026-05-11 |
SAVRN's Notes on medgemma-4b-it
Medical image and text comprehension is what Google trained this for, and the 4B instruction-tuned build is the smallest of three MedGemma variants; the other two are 27B. At 16-bit the weights are 8.6 GB and the run needs 10.3 GB, a sliver of the 192 GB on one MI300X at $1.85 an hour on-demand. At 8-bit the need drops to 5.2 GB, at 4-bit to 2.6 GB. Fit is not the constraint; sessions per card is.
The license reads "other" and access is gated, so a deployment begins with Google's Health AI Developer Foundations terms rather than a standard open license, and those terms come before the files do. Confirm you are taking the build derived from medgemma-4b-pt, and check its context length against your prompts, since our record carries no figure. No host prices are on the Index yet, so your own facility is the priced path.
SAVRN's Notes on Qwen3-VL-4B-Instruct
Feed it an image and a question and it answers in text. At 4.4B parameters the memory figures put it on shared hardware: 2.7 GB needed at 4-bit, 5.3 GB at 8-bit, 10.7 GB for the full 16-bit weights. The cheapest Index setup is one MI300X with 192 GB at $1.85 an hour on-demand, a price set by the card rather than the model, so a single copy leaves most of that card idle.
Apache 2.0 permits commercial use, modification and redistribution, so a fine-tuned version can ship inside a product if the license and copyright notices travel with it and significant changes are stated. Confirm the 262,144-token context length, the ceiling on how much image and text goes into one request, and check that the files you receive match the October 11, 2025 release. Four linked technical reports carry the lineage back to the Qwen-VL paper.
Questions
Which is larger, medgemma-4b-it or Qwen3-VL-4B-Instruct?
Qwen3-VL-4B-Instruct (4.4B parameters) is larger than medgemma-4b-it (4.3B parameters), by the parameter counts their publishers report.
Which is cheaper to run, medgemma-4b-it or Qwen3-VL-4B-Instruct?
At 4-bit, medgemma-4b-it fits on 1x MI300X from $1.85 an hour and Qwen3-VL-4B-Instruct on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use Qwen3-VL-4B-Instruct commercially?
Yes. Qwen3-VL-4B-Instruct 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.