Ling 2.6 Flash
Apr 2026
Ling 2.6 FlashCompare inon
Benchmarks of Ling 2.6 Flash14
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding1
Agents2
Documents1
Legacy · saturated or no longer reported3
No benchmark matches
—per task?
self-host onlyprice
—output speed?
262Kcontext
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-6 Lunalow×2.2—$0.005Function calling only with none · Chat Completions+3
OpenQwen3.8 Flashthinking×4.1—$0.37Memory (4-bit): 64 GB → 108 GB+6BetterGPT-5.6 Lunalow×2.2—$0.01
BetterClaude Haiku 5.5low×3—$0.024Remove sampling parameters+5BetterLing 3.0 FlashthinkingOpen×2.1—$0.028Memory (4-bit): 64 GB → 76 GBModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-6 Lunalow×2.2—$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow×2.2—$0.01Claude Haiku 5.5low×3—$0.024Remove sampling parameters+5Ling 3.0 FlashthinkingOpen×2.1—$0.028Memory (4-bit): 64 GB → 76 GB
MiMo v2.5thinkingOpen×2.3—$0.042Memory (4-bit): 64 GB → 186 GBModelScoreCostPer taskWhat changes
Qwen3.8 FlashthinkingOpen×4.1—$0.37Memory (4-bit): 64 GB → 108 GB+6Qwen3.8 27BxhighOpen×3.5—$1.15max_tokens range · QwenCloud+4
Nemotron 3.5 Super VLthinkingOpen×2.7—$0.19Memory (4-bit): 64 GB → 75 GBTInkling SmallthinkingOpen×2.7—$0.089Context: 262K → 256K+2IK2 Horizon MoVA 36B A4BthinkingOpen×2.6—$1 / $4ModelScoreCostPer taskWhat changes
Ling 3.1 Flash 1019×4.5—$1.49Ling 3.1 FlashOpen×4.2—$1.01Memory (4-bit): 64 GB → 343 GBLing 3.0 Flash VLthinkingOpen×2.5—$0.06 / $0.18Memory (4-bit): 64 GB → 75 GBLing 3.0 Flash FinthinkingOpen×2.3—$0.06 / $0.18Memory (4-bit): 64 GB → 76 GBLing 3.0 FlashthinkingOpen×2.1—$0.028Memory (4-bit): 64 GB → 76 GBLing 2.6 1TOpen+76%——Memory (4-bit): 64 GB → 615 GBRing 2.6 1TOpen+72%—$0.29Context: 262K → 128K+1Ling 3.0 TinyOpen+14%——All inclusionAI models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✗ 48 GB GPU✓ 80 GB GPU · compressed
LicenceMITfree for commercial use
History
5 versionsBenchmarks
all categoriesChat & Knowledge
Humanity's Last Exam6.3#244
AA Omniscience Index-66.1#287
CritPt (physics)0#182
AA Intelligence Index 4.3.2≈9.7#215
AA Omniscience · Accuracy15.6#227
AA-LCR (long context) 1.131.3#222
GPQA Diamond59.3#229
Coding
Terminal-Bench 2.124.3#107
Agents & Tool Use
τ³-Banking · AA2.9#132
τ²-bench · AA86#61
Writing & Language
IFBench (instructions)57.4#94
FAQ
How much does Ling 2.6 Flash cost?›
There is no per-unit API price: self-host only.
Is Ling 2.6 Flash free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Ling 2.6 Flash, and what is it best at?›
#242 of 336 in Chat & Knowledge, #88 of 239 in Coding, #146 of 177 in Agents & Tool Use, #94 of 245 in Writing & Language. Strongest at: Tool calling.
Is Ling 2.6 Flash better than Ling Flash 2.0?›
On AA Intelligence Index 4.3.2, Ling 2.6 Flash scores 9.7 vs 7.8 — higher.
What is the context window of Ling 2.6 Flash?›
262K tokens.
When was Ling 2.6 Flash released? Is there a newer version?›
Released 2026-04-21. It has been superseded by Ling 3.0 Flash.
Is Ling 2.6 Flash open source?›
Yes, the weights are open (MIT licence).
Can I run Ling 2.6 Flash locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Ling 2.6 Flash read images?›
No, it accepts text only.
What is Ling 2.6 Flash?›
Ling 2.6 Flash is an open-weights language model from inclusionAI, released Apr 2026, the successor to Ling Flash 2.0.
What are the best alternatives to Ling 2.6 Flash?›
GPT-6 Luna (better), Qwen3.8 Flash (open weights).