# dolphin-2.9.1-yi-1.5-34b vs Qwen3.6-35B-A3B-abliterated-v4
Source: https://savrn.com/models/compare/dolphin-2-9-1-yi-1-5-34b-vs-qwen3-6-35b-a3b-abliterated-v4
Markdown alternate of the page above; the site index is https://savrn.com/llms.txt

---

| Field | [dolphin-2.9.1-yi-1.5-34b](https://savrn.com/models/dolphin-2-9-1-yi-1-5-34b) dphn/dolphin-2.9.1-yi-1.5-34b | [Qwen3.6-35B-A3B-abliterated-v4](https://savrn.com/models/qwen3-6-35b-a3b-abliterated-v4) Bahushruth/Qwen3.6-35B-A3B-abliterated-v4 |
| --- | --- | --- |
| Publisher | Dolphin | CS |
| Task | Text generation | Text generation |
| Modality | Text | Text |
| Parameters, as reported | 34.4B parameters | 34.7B parameters |
| Architecture | LlamaForCausalLM | Qwen3_5MoeForCausalLM |
| Library | transformers | transformers |
| Context length | 8,192 tokens | 262,144 tokens |
| Repository size | 68.8 GB | 69.3 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) | 82.5 GB | 83.2 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 20.6 GB | 20.8 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 0141cba238d0 | 1c28f5014d7e |
| Downloads reported by the hub | 4.6M | 896.6k |
| Last observed | 2026-10-25 | 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 dolphin-2.9.1-yi-1.5-34b

A 34.4B-parameter model that still fits one rented card: 82.5 GB at 16-bit, 41.3 GB at 8-bit, 20.6 GB at 4-bit, and the cheapest listed setup at every precision is one MI300X with 192 GB at $1.85 an hour on-demand. That makes this one of the few ways to run an unquantized 34B for under two dollars an hour. The 8,192-token context is the limit to design around — it sits well below the 32K windows on the Qwen coder and VL neighbors here — so long-document work should look at those first, and use this for what fits.

The listing records a derivation from 01-ai/Yi-1.5-34B with a LlamaForCausalLM architecture, and the disclosed training mixes — function-calling-chatml, Dolphin-2.9, dolphin-coder — point at code and tool-calling use rather than long-form prose. Apache 2.0 permits commercial use, modification and redistribution with notices kept and changes stated; read the base model's terms alongside. The repo is 24 files totaling 68.8 GB in bfloat16 safetensors, and the files last moved September 8, 2025. No Index host sells it by the token.

SAVRN Research

## SAVRN's Notes on Qwen3.6-35B-A3B-abliterated-v4

83.2 GB at 16-bit, 20.8 GB at 4-bit for a 34.7B-parameter mixture-of-experts with 262,144 tokens of context: the whole thing fits one 192 GB MI300X at $1.85 an hour, with headroom the long-context cache will happily consume. That is the practical news — a quarter-million-token window usually means a multi-GPU bill, and here it does not. A GGUF conversion of the same checkpoint sits in a sibling repository if your stack is llama.cpp rather than vLLM.

What you are actually downloading is Qwen/Qwen3.6-35B-A3B with its refusal direction surgically removed, per the publisher's own table: harmful-prompt refusals drop from roughly 98% to zero, with harmless-prompt refusals unchanged at zero. The refusal policy moves out of the weights and into your gateway, which is where an operator has to own it. Apache 2.0 keeps commercial use open, and no third-party evaluations are reported for this edit, so budget time to regression-test it yourself.

SAVRN Research

## Questions

### Which is larger, dolphin-2.9.1-yi-1.5-34b or Qwen3.6-35B-A3B-abliterated-v4?

Qwen3.6-35B-A3B-abliterated-v4 (34.7B parameters) is larger than dolphin-2.9.1-yi-1.5-34b (34.4B parameters), by the parameter counts their publishers report.

### Which is cheaper to run, dolphin-2.9.1-yi-1.5-34b or Qwen3.6-35B-A3B-abliterated-v4?

At 4-bit, dolphin-2.9.1-yi-1.5-34b fits on 1x MI300X from $1.85 an hour and Qwen3.6-35B-A3B-abliterated-v4 on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

### Can I use dolphin-2.9.1-yi-1.5-34b commercially?

Yes. dolphin-2.9.1-yi-1.5-34b 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 Qwen3.6-35B-A3B-abliterated-v4 commercially?

Yes. Qwen3.6-35B-A3B-abliterated-v4 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

- [dolphin-2.9.1-yi-1.5-34b vs Qwen2.5-32B-Instruct](https://savrn.com/models/compare/dolphin-2-9-1-yi-1-5-34b-vs-qwen2-5-32b-instruct)
- [dolphin-2.9.1-yi-1.5-34b vs Qwen2.5-Coder-32B-Instruct](https://savrn.com/models/compare/dolphin-2-9-1-yi-1-5-34b-vs-qwen2-5-coder-32b-instruct)