Ant GroupLing Flash 2.0
Ling Flash 2.0 is an Ant Group open-weights model, released 17 Sept 2025. Its best configuration ranks #538 of 785 on the BenchLeader Index at 44.4 ±4.6, in the lower half. The ± is the point: 302 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in coding (53) and lowest in long context (33). At $0.248 per million tokens blended it is cheaper than most ranked models. Output speed of 4 tokens per second puts it in the slowest quarter, with a first answer in 2.5 s. Last measured 22 Oct 2026.
- Blended price
- $0.248/M
- $0.140 in · $0.570 out
- Output speed
- 4 tok/s
- measured by Artificial Analysis
- First answer
- 2.49 s
- first token 2.49 s
- Context
- 131k
- Full answer
- 122 s
- median, reasoning included
- Cost per run
- –
- Released
- 17 Sept 2025
- Overall index44
- Reasoning44
- Coding53
- Agents & tools43
- Maths51
- Knowledge33
- Instruction following38
- Human preference51
- Long context33
- Composite38
Versions
Ant Group has shipped 3 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Ling 3.1 Flash | 1 Oct 2026 | – | – |
| Ling 3.0 Flash | 4 Aug 2026 | 52.8 | #303 |
| Ling Flash 2.0this page | 17 Sept 2025 | 44.4 | #538 |
How its index has moved
9 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Hard Prompts | 1365#202 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 65.7%#339 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 6.2%#401 | Artificial Analysis | |
| CritPt | 0.0%#315 | Artificial Analysis |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Coding | 1409#187 | Arena (LMArena) |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Terminal-Bench Hard | 10.6%#221 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 20.8%#329 | Artificial Analysis |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Maths | 1345#204 | Arena (LMArena) |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-Omniscience | -62.7#483 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 14.1%#443 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 10.5%#408 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| IFBench | 34.4%#319 | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1316#221 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1343#208 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1268#249 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1312#235 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1326#230 | Arena (LMArena) |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR | 17.7%#445 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA Intelligence Index v4.3.2 | 7.8#457 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0002 | – | 1.2 min |
| Summarize a 30-page report | 12,000 / 600 | $0.0020 | – | 2.4 min |
| Code edit | 6,000 / 1,500 | $0.0017 | – | 6.0 min |
| Agentic coding session | 60,000 / 4,000 | $0.011 | – | 16.0 min |
| Structured extraction | 2,000 / 200 | $0.0004 | – | 50.3 s |
More from Ant Group
See also
Questions people ask
- What are Ling Flash 2.0's benchmark scores?
- As of 24 Oct 2026, Ling Flash 2.0 scores 44.4 ±4.6 on the BenchLeader Index, strongest in coding (53) and weakest in long context (33). That rests on 10 independent benchmark results, and every score on this page links to its publisher.
- How does Ling Flash 2.0 rank?
- Ling Flash 2.0 is #538 of 785 ranked configurations on 24 Oct 2026. 302 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Ling Flash 2.0 cost per 1M tokens?
- As of 24 Oct 2026, Ling Flash 2.0 costs $0.140 per 1M input tokens and $0.570 per 1M output tokens, $0.248 blended (three input tokens per output token).
- When was Ling Flash 2.0 released?
- Ant Group released Ling Flash 2.0 on 17 Sept 2025.
Data as of 24 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Ling Flash 2.0: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/ling-flash-2-0.