Qwen3.8 Flash Next
A lightweight model variant within Alibaba's Qwen3 series, designed for efficient inference scenarios. It supports community quantization and local deployment.
Core Facts
The approximately 20:1 ratio between Qwen3 Next Flash's 125 billion total parameters and 6 billion activated parameters represents a fairly aggressive sparsity design.
In an RTX 5070 (12GB VRAM) configuration, the local Qwen3 9B model's output speed reached approximately 70–76 tokens/second.
Timeline (last 90 days)
Tested on a machine with an AMD 9700 + 32GB of VRAM + 64GB of memory, the model can run a 56K context.
The quantized Coder model deployed with this approach has limited text generation quality and is suitable for low-cost experimentation or as a free token source for agents.
According to benchmark data from Hugging Face, the IQ3_S tier scores an average of 93.26, close to the full-precision base model's score of 93.12.
Qwen3.8 Flash Next has 125B total parameters, activates only approximately 6B per token, supports a native context of 260,000 tokens expandable to 1 million, and its training cost is reportedly only one-ninth that of the previous generation.
Strata enables a single GPU with 12 GB of VRAM to smoothly run Qwen3.8 Flash Next, a large model with 125 billion activated parameters and an overall size of 176B.
Qianwen officially placed Qwen3.8 Flash Next into a subscription plan to sell API access externally, while the 27B version has not yet been sold separately.
Actual testing of two quantization models: IQ2_XXS achieves 48 tokens/s for short Q&A and 63 tokens/s for long text; IQ3_XXS achieves 30 tokens/s for short Q&A and 39 tokens/s for long text, with VRAM usage of approximately 16.44 GB, activating 10,138 experts—IQ2 is about 60% faster than IQ3.
When extending the context from 192,000 to 256,000, the expert cache dropped from 16.44 GB to 13.66 GB and KV increased from 4.31 GB to 7 GB, with almost zero speed cost, because KV is fully in VRAM and the expert cache automatically balances.
On a single RX 7900 XTX 24G GPU, the Strata engine can fully run the approximately 125B-parameter Qwen3.8 Flash Next MoE model, supporting both text inference and image recognition.
Strata is an open-source inference runtime capable of running the 125-billion-parameter Qwen3.8 Flash Next model on an RTX 5070 (12 GB VRAM) at an output speed of approximately 90 tokens per second.
23 more timeline events
All Facts (20)
The approximately 20:1 ratio between Qwen3 Next Flash's 125 billion total parameters and 6 billion activated parameters represents a fairly aggressive sparsity design.
70%VerifiedIn an RTX 5070 (12GB VRAM) configuration, the local Qwen3 9B model's output speed reached approximately 70–76 tokens/second.
70%UnverifiedQwen3.8 Flash Next supports two structured output modes: function calling and constrained generation.
60%UnverifiedThe quantized Coder model deployed with this approach has limited text generation quality and is suitable for low-cost experimentation or as a free token source for agents.
50%UnverifiedTested on a machine with an AMD 9700 + 32GB of VRAM + 64GB of memory, the model can run a 56K context.
50%UnverifiedAccording to benchmark data from Hugging Face, the IQ3_S tier scores an average of 93.26, close to the full-precision base model's score of 93.12.
50%UnverifiedQwen3.8 Flash Next has 125B total parameters, activates only approximately 6B per token, supports a native context of 260,000 tokens expandable to 1 million, and its training cost is reportedly only one-ninth that of the previous generation.
50%UnverifiedStrata enables a single GPU with 12 GB of VRAM to smoothly run Qwen3.8 Flash Next, a large model with 125 billion activated parameters and an overall size of 176B.
50%UnverifiedWhen extending the context from 192,000 to 256,000, the expert cache dropped from 16.44 GB to 13.66 GB and KV increased from 4.31 GB to 7 GB, with almost zero speed cost, because KV is fully in VRAM and the expert cache automatically balances.
50%UnverifiedOn a single RX 7900 XTX 24G GPU, the Strata engine can fully run the approximately 125B-parameter Qwen3.8 Flash Next MoE model, supporting both text inference and image recognition.
50%UnverifiedActual testing of two quantization models: IQ2_XXS achieves 48 tokens/s for short Q&A and 63 tokens/s for long text; IQ3_XXS achieves 30 tokens/s for short Q&A and 39 tokens/s for long text, with VRAM usage of approximately 16.44 GB, activating 10,138 experts—IQ2 is about 60% faster than IQ3.
50%UnverifiedStrata is an open-source inference runtime capable of running the 125-billion-parameter Qwen3.8 Flash Next model on an RTX 5070 (12 GB VRAM) at an output speed of approximately 90 tokens per second.
50%UnverifiedIn Qwen3.8 Flash Next, each layer has a router that selects 10 active experts out of 512, plus one shared expert that always runs.
50%UnverifiedThe Qwen3.8 Flash Next model has 48 layers in total, constructed by repeating "3 gated DeltaNet layers + 1 sparse injection" 12 times, containing approximately 245,000 expert instances in total, with traditional attention cache retained in only 12 layers.
50%UnverifiedThe naming jump from Qwen3.8 to Qwen4 may imply architectural changes larger than a minor version update.
50%UnverifiedA Qwen Flash model released by Alibaba at the end of August also places 51 billion n-gram parameters in system memory for execution.
50%UnverifiedQwen3.8 Flash Next adopts a hybrid architecture whose core technology is Gated DeltaNet, combining linear attention with standard full attention
50%UnverifiedQwen3.8 Flash Next has a native context window of 262K tokens, which can be extended to 1 million tokens using YaRN technology
50%UnverifiedQwen3.8 Flash Next scored 84.5 on the Android World benchmark and 91.9/95.7 on Live Code Bench (both are official figures)
50%UnverifiedQwen3.8 Flash Next has 125 billion total parameters, with only approximately 6 billion parameters activated per token during inference
50%