Ctrl K
Back to models

Alibaba

Qwen3.8-Flash-Next

  • Released
  • Open weights
  • Qwen Community License 1.0
Compare
Intelligence Index

Reasoning setting

39.8#22 of 83measured models
Measured · Artificial Analysis v4.3.2
API price from
$0.15 in / $0.47 out / 1M tokOpenRouter · Alibaba
Context window
262.14K tok

Performance profile

ReasoningRank #13 of 107
92.3%
GPQA Diamond
CodingRank #32 of 94
50.6%
SciCode
Cost efficiencyRank #28 of 67
107.0pts / $
Intelligence ÷ eval cost

Capabilities

Context window

Modalities

Reasoning modes

Integration

Reasoning trade-offs

Source (opens in a new tab)
Intelligence Index v4.3.2
Qwen3.8-Flash-Next
Cost per task (USD)
Qwen3.8-Flash-NextReasoning
Intelligence Index
39.8
Cost / task
$0.3722
Evaluation time
27.1m

Pricing & access

API route
OpenRouter · AlibabaAPI
Tier & contextStandard

USD per 1M tokens

Input
$0.15
Output
$0.47
Cached input
$0.016

Run locally

GGUF
Context
Estimated memory

~122.4 GiB

RAM + VRAM

Estimate assumptions

One text chat at the selected context, with a 16-bit cache and app allowance. Operating system and image processing are extra; optimized caches can use less.

128 GiB RAMwithout GPU offload

Memory budgetOne text chat
Weights
111 GiB
Context + state
0.3 GiB16-bit cache
App allowance
11.1 GiB
Download
119.2 GB3 files · keep together+0.9 GB for images (opens in a new tab)
Disk space
~138 GBIncludes image file
Available software
Ollama (opens in a new tab)LM Studio (opens in a new tab)Jan (opens in a new tab)KoboldCpp (opens in a new tab)llama.cpp (opens in a new tab)
Get GGUF (opens in a new tab)
Set up in Ollama · Q4_K_M · 8K context

Save the following as Modelfile. The context matches the estimate above.

FROM hf.co/lmstudio-community/Qwen3.8-Flash-Next-GGUF:Q4_K_M
PARAMETER num_ctx 8192

Then run these commands in your terminal:

ollama create llmcurrent-local -f Modelfile
ollama run llmcurrent-local

Ollama import instructions (opens in a new tab) · For LM Studio or Jan, open the app’s model search, enter lmstudio-community/Qwen3.8-Flash-Next-GGUF, choose Q4_K_M, and set context to 8,192 tokens in the app’s model settings. Software links above open each app’s setup guide.

Images need extra memoryvLLM setup (opens in a new tab)

Benchmarks

Full leaderboardSource (opens in a new tab)

Humanity’s Last Exam

Frontier reasoning

Qwen3.8-Flash-Next · Reasoning#31 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 28MiniMax-M3Reasoning39.0%
  6. 29GPT-6 LunaMax38.5%
  7. 30Agnes 3.0 FlashReasoning38.5%
  8. 31Qwen3.8-Flash-NextReasoning38.0%
  9. 32Agnes 2.5 Pro BetaReasoning37.5%
  10. 33Motif 3Reasoning37.0%
  11. 34Qwen3.7 PlusReasoning35.6%

Peers use their best measured result on this benchmark.

Alternatives

Each model at its highest-scoring measured setting · Intelligence Index v4.3.2 and its evaluation cost per task.