Benchmarks of JT-4.1 Flash 236B A21B16
BenchmarkByScoreRank
Overall1
Knowledge3
Science3
Coding2
Agents3
Multimodal1
Documents2
Legacy · saturated or no longer reported1
No benchmark matches
—per task?
API price not publicprice
—output speed?
262Kcontext
Rankingin thinking mode?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterMiMo v2.6 FlashthinkingOpen+12%—$0.062Pass back reasoning_content+3OpenMiMo v2.6 Prothinking+37%—$0.13Pass back reasoning_content+3
BetterGPT-6 Lunamax+12%—$0.068Function calling only with none · Chat Completions+3
BetterClaude Haiku 5.5high+12%—$0.079Remove sampling parameters+5BetterGPT-6.1 Sollow+24%—$0.13Remove unsupported parameters+5ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
MiMo v2.6 FlashthinkingOpen+12%—$0.062Pass back reasoning_content+3GPT-6 Lunamax+12%—$0.068Function calling only with none · Chat Completions+3Claude Haiku 5.5high+12%—$0.079Remove sampling parameters+5GPT-6.1 Sollow+24%—$0.13Remove unsupported parameters+5MiMo v2.6 ProthinkingOpen+37%—$0.13Pass back reasoning_content+3ModelScoreCostPer taskWhat changes
MiMo v2.6 ProthinkingOpen+37%—$0.13Pass back reasoning_content+3
DeepSeek V4.1 Flash 1017maxOpen+33%—$0.26Pass back reasoning_content · Chat Completions+4
GLM 5.3maxOpen+32%—$2.01No image input+4
Step 5 PreviewthinkingOpen+29%—$1.03web_search not available · Chat Completions+2
Kimi K3maxOpen+29%—$2.00Sampling parameters are fixed+3ModelScoreCostPer taskWhat changes
CJT 35B Flash-45%——Context: 262K → 256K+1CJT Mini-64%——Context: 262K → 128K+1All China Mobile models →Benchmarks
all categoriesChat & Knowledge
Humanity's Last Exam43.6#28
AA Omniscience Index20.6#25
CritPt (physics)14#47
AA Intelligence Index 4.3.233.9#45
AA Omniscience · Accuracy49.4#24
AA-LCR (long context) 1.178.3#61
GDP.pdf · AA all-pass8.6#64
Coding
SciCode52#42
Agents & Tool Use
AutomationBench-AA46.5#40
GDPval-AA (work tasks) 2.141.2#46
AA-Briefcase 1.1 · Elo1,089.3#49
FAQ
How much does JT-4.1 Flash 236B A21B cost?›
There is no per-unit API price: API price not public.
Is JT-4.1 Flash 236B A21B free to use?›
There is no free API.
How good is JT-4.1 Flash 236B A21B, and what is it best at?›
#28 of 334 in Chat & Knowledge, #46 of 132 in Coding, #40 of 133 in Agents & Tool Use, #85 of 124 in Vision & Documents. Strongest at: Reasoning, Long context, Knowledge.
Is JT-4.1 Flash 236B A21B better than Gemini 3.6 Flash?›
On AA Intelligence Index 4.3.2, JT-4.1 Flash 236B A21B scores 33.9 vs 34 — about the same.
What is the context window of JT-4.1 Flash 236B A21B?›
262K tokens.
When was JT-4.1 Flash 236B A21B released? Is there a newer version?›
Released 2026-07-09. It is the current version in its line.
Is JT-4.1 Flash 236B A21B open source?›
No, JT-4.1 Flash 236B A21B is a closed model available through an API. The closest open-weights option here is MiMo v2.6 Pro.
Can JT-4.1 Flash 236B A21B read images?›
It accepts text and images as input.
What is JT-4.1 Flash 236B A21B?›
JT-4.1 Flash 236B A21B is a closed language model from China Mobile, released Jul 2026.
What are the best alternatives to JT-4.1 Flash 236B A21B?›
MiMo v2.6 Flash (better), MiMo v2.6 Pro (open weights).