OSAIM
Open Source AI Models

Qwen 3 235B (A22B)

The flagship Qwen 3 release: a 235B-total MoE with 22B active parameters per token. Competitive with DeepSeek V3 and Llama 4 Maverick on reasoning benchmarks while being smaller total. Apache 2.0 — one of the most permissively licenced frontier-class models.

Parameters
235B
Context length
128K
Modality
text
Released
2025-04-29

Memory & hardware

VRAM (fp16)
470 GB
VRAM (Q4)
141 GB
Recommended
4× H100 80GB at fp8
Quantizations
fp16, fp8, q8_0, q4_k_m

License: Apache 2.0

SPDX
Apache-2.0
Commercial use
Yes
Modification
Yes
Redistribution
Yes

Benchmarks

ArenaHard
93.2
MATH
91.2
HumanEval
90.9
IFEval
87.9
MMLU
87.1
MMLU-Pro
79.1
GPQA
63.5
SWE-bench Verified
39.5
Benchmarks last verified 2026-07-02.

Hosted inference pricing

USD per million tokens.

ProviderInputOutput
together$0.60$0.60
deepinfraCheapest$0.40$1.40
Pricing last verified 2026-07-02. Providers update rates frequently; confirm before integrating.

Run it yourself

Drop-in commands for the three most common open-source inference paths. The Ollama tag is a best-effort match against the registry; verify the size variant before pulling.

Run Qwen 3 235B (A22B) locally
No official Ollama registry tag for this model — use transformers or vLLM below.
vLLM (production)
vllm serve Qwen/Qwen3-235B-A22B
High-throughput hosted inference; one command to expose an OpenAI-compatible HTTP server.
Transformers (Python)
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-235B-A22B")
model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen3-235B-A22B", device_map="auto", torch_dtype="auto"
)
Direct PyTorch usage. Pin a torch / cuda version that matches your GPU.
Hugging Face ID: Qwen/Qwen3-235B-A22B

Related models

Same family or similar size — useful when shopping around.

Qwen2.5 72B Instruct
72B

The flagship Qwen 2.5 release. Competes with Llama 3.1 405B on many benchmarks at one-fifth the parameter count. Note the 72B specifically uses the Qwen License (commercial use up to 100M MAU) — the smaller Qwen2.5 sizes are Apache 2.0.

Context
128K
License
qwen
VRAM Q4
43.2 GB
DeepSeek Coder V2
236B

Coding-focused MoE model with 21B active parameters out of 236B total. Supports 338 programming languages with strong performance across mainstream stacks (Python, TypeScript, Go, Rust, Java, C++) and competent results on niche languages where most open models falter. The DeepSeek licence applies — commercial use permitted with some application restrictions.

Context
128K
License
deepseek
VRAM Q4
141.6 GB
Grok 2
300B

xAI's second open-weights release, Apache 2.0. ~300B mixture-of-experts. xAI's pattern of open-sourcing the previous frontier when a new one ships continues from Grok 1. Competitive with GPT-4-class chat quality at release; today useful mainly as a research artefact given the compute needed to run it.

Context
131K
License
apache-2-0
VRAM Q4
180 GB
Grok 1
314B

xAI's first open-weights release: a 314B-parameter mixture-of-experts model. Apache 2.0 licensed. Largely a research artefact at this size — most users will run smaller models for production — but useful as a permissively-licensed reference for MoE research.

Context
8K
License
apache-2-0
VRAM Q4
188.4 GB
Mixtral 8×22B Instruct
141B

Scaled-up Mixtral with 22B-parameter experts. ~39B active parameters out of 141B total. Strong long-context performance and competitive coding scores. Apache 2.0 makes it attractive for self-hosting where the licence terms of Llama 3 are a non-starter.

Context
66K
License
apache-2-0
VRAM Q4
84.6 GB
DBRX Instruct
132B

Databricks' 132B mixture-of-experts — 16 experts, 4 active per token (36B active params). Trained on 12T tokens on Mosaic infrastructure and released under the Databricks Open Model Licence. DBRX was best-in-class on release; now beaten by Llama 3.3 70B and Qwen 2.5 72B on most benchmarks, but retains value as a well-documented MoE reference.

Context
33K
License
dbrx-open
VRAM Q4
79.2 GB