SAVRN Model Hub · Comparisons
OTel-2.0-LLM-31B-IT vs Qwen3-30B-A3B
OTel-2.0-LLM-31B-IT has 31.3B parameters and Qwen3-30B-A3B has 30.5B parameters; both are released under Apache License 2.0; at 16-bit, OTel-2.0-LLM-31B-IT needs about 75.1 GB (1x MI300X from $1.85 an hour) and Qwen3-30B-A3B about 73.3 GB (1x MI300X from $1.85 an hour).
| Field | OTel-2.0-LLM-31B-IT farbodtavakkoli/OTel-2.0-LLM-31B-IT | Qwen3-30B-A3B Qwen/Qwen3-30B-A3B |
|---|---|---|
| Publisher | Farbod Tavakkoli | Qwen |
| Task | Text generation | Text generation |
| Modality | Text | Text |
| Parameters, as reported | 31.3B parameters | 30.5B parameters |
| Architecture | Gemma4ForConditionalGeneration | Qwen3MoeForCausalLM |
| Library | transformers | transformers |
| Context length | 262,144 tokens | 40,960 tokens |
| Repository size | 126.8 GB | 61.1 GB |
| Artifact formats | safetensors | safetensors |
| License | apache-2.0 | apache-2.0 |
| Access | Open weights, no gate | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 75.1 GB | 73.3 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 18.8 GB | 18.3 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 6d425c390399 | ad44e777bcd1 |
| Downloads reported by the hub | 7.2M | 1.8M |
| 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.
SAVRN's Notes on OTel-2.0-LLM-31B-IT
Post-trained from Gemma 4 31B-IT on roughly 440 billion telecom tokens, OTel-2.0-LLM-31B-IT is aimed at network operations, standards interpretation, configuration help and retrieval-backed answering. Its 31.3B parameters need 75.1 GB at 16-bit, 37.5 GB at 8-bit and 18.8 GB at 4-bit, and all three fit the 192 GB of one MI300X, the cheapest Index setup at $1.85 an hour. The 262,144-token context matters when the input is a standards document rather than a ticket.
Apache 2.0 allows commercial deployment, modification and redistribution, with the notice and change-statement duties kept. Two things to verify. The repository is 39 files and 126.8 GB, about twice the 62.5 GB of 16-bit weights, so confirm what those files hold before pulling them. And because Farbod Tavakkoli post-trained this from a Gemma 4 base, confirm the terms that travel with the base weights fit your plan; no Index host prices are listed.
SAVRN's Notes on Qwen3-30B-A3B
Start with the memory figure: 73.3 GB at 16-bit for a 30.5B-parameter model built from 128 experts. That fits on one MI300X with 192 GB at $1.85 an hour on demand, the cheapest listing on the Index, with room left for the 40,960-token context. Quantize to 8-bit and the footprint is 36.6 GB; at 4-bit it is 18.3 GB. The weight files total 61 GB, so plan storage and load time around that.
Apache 2.0 allows commercial use, modification and redistribution; keep the notices, state significant changes, and contributors give an express patent grant. Weigh the hourly card against DeepInfra's Index price of $0.12 in and $0.50 out per million tokens; token volume decides which wins. It derives from Qwen3-30B-A3B-Base, so the base is available for your own post-training, and the publisher describes a thinking mode and a non-thinking mode in one model, worth testing on your workload.
Questions
Which is larger, OTel-2.0-LLM-31B-IT or Qwen3-30B-A3B?
OTel-2.0-LLM-31B-IT (31.3B parameters) is larger than Qwen3-30B-A3B (30.5B parameters), by the parameter counts their publishers report.
Which is cheaper to run, OTel-2.0-LLM-31B-IT or Qwen3-30B-A3B?
At 4-bit, OTel-2.0-LLM-31B-IT fits on 1x MI300X from $1.85 an hour and Qwen3-30B-A3B on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use OTel-2.0-LLM-31B-IT commercially?
Yes. OTel-2.0-LLM-31B-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 Qwen3-30B-A3B commercially?
Yes. Qwen3-30B-A3B 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.