Ling 3.1 Flash vs Qwen3.8 Max

Change models
Ling 3.1 FlashInclusionAI
IndexAll categories, out of 100.
65.8 ±7.3 best
65.1 ±2.5
GapIndex points below the top.
Top
−0.7 · no clear gap
CoverageIndex weight that has results.
70%
95% best
SpeedOutput tokens per second.
84/s best
41/s
EloLMArena rating and rank.
—
1485
ReleasedMonth of release.
2026-09
2026-08
Input / 1MUS dollars per 1M input tokens.
$0.045
—
Output / 1MUS dollars per 1M output tokens.
$0.135
—
1K chats2K tokens in, 500 out, each.
$0.158
—
1K documents20K tokens in, 1K out, each.
$1.04
—
1K agent tasks200K tokens in, 20K out, each.
$11.70
—
ContextMaximum tokens in one request.
1M
1M
AgenticMulti-step tasks with tools.
66.2 best
63.3
CodingCode writing and repair.
66.5 best
63.8
ReasoningLogic problems and puzzles.
67.6 best
66.4
MultimodalTasks with images and text.
—
66.8
KnowledgeFacts and expert knowledge.
62.5 best
62.3
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
71.1
MathMath problems.
—
67.2

Benchmarks

6 shared
Only Ling 3.1 Flash: 12Only Qwen3.8 Max: 108
Ling 3.1 Flash
Qwen3.8 Max
81.2%31 / 86
86.6% best15 / 86
68.7% best10 / 83
67.7%11 / 83
37.4%39 / 79
43.6% best31 / 79
DeepSWEAgentic
59.7% best44 / 61
56.6%48 / 61
52.5% best8 / 33
27.3%31 / 33
83.3% best2 / 16
81.9%3 / 16