Skip to content

InclusionAI · Model profile

Ling-flash-2.0

Intelligence, Performance & Price Analysis

Source: Canonical slug: ling-flash-2-0 · Canonical model registry at build time

Routing verdict

Capable everyday model

Good for routine tasks; route complex reasoning and premium workloads to stronger models.

Ling-flash-2.0 summary metrics

Intelligence
10 scoreLing-flash-2.0 scores 10 (estimated) on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 7).
Speed
65.9 output tokens/secLing-flash-2.0 generates output at 65.9 tokens per second (based on the median across providers serving the model), which is below average compared to other open weight non-reasoning models of similar size (median: 81.5 t/s).
Latency
2.23s TTFTLing-flash-2.0 has a time to first token (TTFT) of 2.23s (based on the median across providers serving the model), which is somewhat higher than average compared to other open weight non-reasoning models of similar size (median: 1.59s).
Input Price
$0.14 / 1M tokensLing-flash-2.0 costs $0.14 per 1M input tokens (very competitive, median: $0.53) and $0.57 per 1M output tokens (very competitive, median: $1.05), based on the median across providers serving the model.
Output Price
$0.57 / 1M tokensLing-flash-2.0 costs $0.14 per 1M input tokens (very competitive, median: $0.53) and $0.57 per 1M output tokens (very competitive, median: $1.05), based on the median across providers serving the model.

Assessment

Routing verdict and task fit

Ling-flash-2.0 scores 10 (estimated) on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 7). Ling-flash-2.0 generates output at 65.9 tokens per second (based on the median across providers serving the model), which is below average compared to other open weight non-reasoning models of similar size (median: 81.5 t/s). Ling-flash-2.0 costs $0.14 per 1M input tokens (very competitive, median: $0.53) and $0.57 per 1M output tokens (very competitive, median: $1.05), based on the median across providers serving the model.
Task fit assessment
WorkloadRatingNotes
Complex reasoning & agentic workflowsViableIntelligence score 10 handles routine reasoning but may struggle with open-ended agentic tasks.
High-volume chat & customer-facingOptimalOutput speed 65.9 tokens/sec is adequate for chat.
Latency-sensitive applicationsOptimalTTFT 2.23s — adequate latency for most interactive use cases.
Cost-sensitive pipelinesOptimalOutput pricing at $0.57 is reasonable for moderate volume.

Evidence

Evidence charts

16 published comparison charts for this profile. Ling-flash-2.0 is marked with a diamond and bold label; each chart has a table view and lists its source and retrieval date.

AA-Omniscience Index

AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis

17 models in the published order. Ling-flash-2.0: -63.4, row 16 of 17 in the published order.
  1. Claude Fable 5 (with fallback)40.1
  2. Gemini 3.5 Flash22.7
  3. GPT-5.5 (xhigh)20.1
  4. Grok 4.3 (high)18.3
  5. Kimi K2.66.4
  6. Muse Spark4.1
  7. GLM-5.2 (max)4
  8. MiniMax-M31.4
  9. Nemotron 3 Ultra-0.833
  10. DeepSeek V4 Pro (max)-10
  11. Devstral 2-46.2
  12. gpt-oss-120b (high)-50
  13. Llama 3.3 70B-51.9
  14. Llama 4 Scout-52.4
  15. Qwen3 Coder Next-60.7
  16. Ling-flash-2.0 (this model)-63.4
  17. Ling 2.6 Flash-65.7

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Artificial Analysis Intelligence Index by Open Weights / Proprietary

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis

17 models in the published order. Ling-flash-2.0: 9.7, row 16 of 17 in the published order.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. gpt-oss-120b (high)23.8
  12. Qwen3 Coder Next21.1
  13. Devstral 219.2
  14. Ling 2.6 Flash14.1
  15. Llama 4 Scout10
  16. Ling-flash-2.0 (this model)9.7
  17. Llama 3.3 70B9.4

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis

17 models in the published order. Ling-flash-2.0: 9.7, row 16 of 17 in the published order.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. gpt-oss-120b (high)23.8
  12. Qwen3 Coder Next21.1
  13. Devstral 219.2
  14. Ling 2.6 Flash14.1
  15. Llama 4 Scout10
  16. Ling-flash-2.0 (this model)9.7
  17. Llama 3.3 70B9.4

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Artificial Analysis Openness Index: Score

Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open) · Evaluation results measured independently by Artificial Analysis

11 models in the published order. Ling-flash-2.0: 44.4, row 3 of 11 in the published order.
  1. Nemotron 3 Ultra83.3
  2. DeepSeek V4 Pro (max)50
  3. Ling-flash-2.0 (this model)44.4
  4. GLM-5.2 (max)44.4
  5. Qwen3 Coder Next41.7
  6. gpt-oss-120b (high)38.9
  7. Ling 2.6 Flash38.9
  8. Llama 3.3 70B38.9
  9. Kimi K2.633.3
  10. MiniMax-M333.3
  11. Llama 4 Scout27.8

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Intelligence

Artificial Analysis Intelligence Index · Higher is better · Evaluation results measured independently by Artificial Analysis

12 models in the published order. Ling-flash-2.0: 9.7, row 12 of 12 in the published order.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. gpt-oss-120b (high)23.8
  12. Ling-flash-2.0 (this model)9.7

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Output Speed

Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis

16 models in the published order. Ling-flash-2.0: 66 t/s, row 16 of 16 in the published order.
  1. gpt-oss-120b (high)314 t/s
  2. Nemotron 3 Ultra250 t/s
  3. GLM-5.2 (max)218 t/s
  4. Gemini 3.5 Flash192 t/s
  5. Ling 2.6 Flash177 t/s
  6. Grok 4.3 (high)164 t/s
  7. Llama 4 Scout109 t/s
  8. MiniMax-M399 t/s
  9. Qwen3 Coder Next89 t/s
  10. GPT-5.5 (xhigh)88 t/s
  11. Llama 3.3 70B86 t/s
  12. Devstral 276 t/s
  13. Kimi K2.676 t/s
  14. DeepSeek V4 Pro (max)72 t/s
  15. Claude Fable 5 (with fallback)71 t/s
  16. Ling-flash-2.0 (this model)66 t/s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Output speed context
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.

Speed

Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis

11 models in the published order. Ling-flash-2.0: 66 t/s, row 11 of 11 in the published order.
  1. gpt-oss-120b (high)314 t/s
  2. Nemotron 3 Ultra250 t/s
  3. GLM-5.2 (max)218 t/s
  4. Gemini 3.5 Flash192 t/s
  5. Grok 4.3 (high)164 t/s
  6. MiniMax-M399 t/s
  7. GPT-5.5 (xhigh)88 t/s
  8. Kimi K2.676 t/s
  9. DeepSeek V4 Pro (max)72 t/s
  10. Claude Fable 5 (with fallback)71 t/s
  11. Ling-flash-2.0 (this model)66 t/s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Output speed context
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.

End-to-End Response Time

Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better · Evaluation results measured independently by Artificial Analysis

16 models in the published order. Ling-flash-2.0: 7.59 s, row 7 of 16 in the published order.
  1. Ling 2.6 Flash2.82 s
  2. Llama 4 Scout4.61 s
  3. Llama 3.3 70B5.85 s
  4. Qwen3 Coder Next5.65 s
  5. Devstral 26.55 s
  6. gpt-oss-120b (high)1.59 s
  7. Ling-flash-2.0 (this model)7.59 s
  8. Nemotron 3 Ultra2 s
  9. GLM-5.2 (max)2.3 s
  10. Grok 4.3 (high)3.05 s
  11. MiniMax-M35.05 s
  12. Gemini 3.5 Flash2.6 s
  13. Kimi K2.66.56 s
  14. DeepSeek V4 Pro (max)6.97 s
  15. GPT-5.5 (xhigh)5.68 s
  16. Claude Fable 5 (with fallback)7.07 s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Latency: Time To First Answer Token

Seconds to first answer token received · Accounts for reasoning model 'thinking' time · Evaluation results measured independently by Artificial Analysis

16 models in the published order. Ling-flash-2.0: 0 s, row 6 of 16 in the published order.
  1. Llama 4 Scout0 s
  2. Devstral 20 s
  3. Ling 2.6 Flash0 s
  4. Llama 3.3 70B0 s
  5. Qwen3 Coder Next0 s
  6. Ling-flash-2.0 (this model)0 s
  7. gpt-oss-120b (high)6.37 s
  8. Nemotron 3 Ultra9.09 s
  9. GLM-5.2 (max)9.19 s
  10. Grok 4.3 (high)0 s
  11. MiniMax-M320.19 s
  12. Gemini 3.5 Flash0 s
  13. Kimi K2.658.43 s
  14. DeepSeek V4 Pro (max)60.99 s
  15. GPT-5.5 (xhigh)0 s
  16. Claude Fable 5 (with fallback)0 s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Latency considerations
Latency includes thinking / reasoning time when applicable. Lower is better for interactive use cases.

Context Window

Context window: tokens limit · Higher is better · Evaluation results measured independently by Artificial Analysis

17 models in the published order. Ling-flash-2.0: 128k, row 16 of 17 in the published order.
  1. Llama 4 Scout10M
  2. Gemini 3.5 Flash1M
  3. Claude Fable 5 (with fallback)1M
  4. GLM-5.2 (max)1M
  5. DeepSeek V4 Pro (max)1M
  6. Grok 4.3 (high)1M
  7. MiniMax-M31M
  8. GPT-5.5 (xhigh)922k
  9. Nemotron 3 Ultra262.1k
  10. Muse Spark262.1k
  11. Ling 2.6 Flash262.1k
  12. Kimi K2.6256k
  13. Qwen3 Coder Next256k
  14. Devstral 2256k
  15. gpt-oss-120b (high)131.1k
  16. Ling-flash-2.0 (this model)128k
  17. Llama 3.3 70B128k

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Context window sizing
Larger context windows enable complex multi-turn tasks but may increase latency and cost for long inputs.

Cost per Intelligence Index Task

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better · Evaluation results measured independently by Artificial Analysis

13 models in the published order. Ling-flash-2.0 is not marked in this published set of 13 models.
  1. Llama 4 Scout0.000824
  2. DeepSeek V4 Pro (max)0.0063
  3. gpt-oss-120b (high)0.00454
  4. Llama 3.3 70B0.0012
  5. MiniMax-M30.0139
  6. Grok 4.3 (high)0.0101
  7. Nemotron 3 Ultra0.0209
  8. Qwen3 Coder Next0.0184
  9. Kimi K2.60.0216
  10. GLM-5.2 (max)0.0268
  11. Gemini 3.5 Flash0.0884
  12. GPT-5.5 (xhigh)0.107
  13. Claude Fable 5 (with fallback)0.367

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Cost per Task

Weighted average cost (USD) per Intelligence Index task · Lower is better · Evaluation results measured independently by Artificial Analysis

10 models in the published order. Ling-flash-2.0 is not marked in this published set of 10 models.
  1. DeepSeek V4 Pro (max)$0.045
  2. gpt-oss-120b (high)$0.061
  3. MiniMax-M3$0.125
  4. Grok 4.3 (high)$0.139
  5. Nemotron 3 Ultra$0.244
  6. Kimi K2.6$0.346
  7. GLM-5.2 (max)$0.370
  8. Gemini 3.5 Flash$0.586
  9. GPT-5.5 (xhigh)$0.863
  10. Claude Fable 5 (with fallback)$2.75

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Cost to Run Artificial Analysis Intelligence Index

Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index · Evaluation results measured independently by Artificial Analysis

13 models in the published order. Ling-flash-2.0 is not marked in this published set of 13 models.
  1. Llama 4 Scout2.7
  2. Llama 3.3 70B2.7
  3. gpt-oss-120b (high)3.5
  4. DeepSeek V4 Pro (max)7.2
  5. MiniMax-M314.5
  6. Qwen3 Coder Next38.5
  7. Grok 4.3 (high)12.5
  8. Nemotron 3 Ultra33.5
  9. GLM-5.2 (max)32.9
  10. Kimi K2.625
  11. Gemini 3.5 Flash114
  12. GPT-5.5 (xhigh)131.8
  13. Claude Fable 5 (with fallback)507.6

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens) · Evaluation results measured independently by Artificial Analysis

15 models; segments: Cache Hit, Input, Output. Ling-flash-2.0: $0.710, row 2 of 15 in the published order.
  • Cache Hit
  • Input
  • Output
  1. Ling 2.6 Flash$0.420
  2. Ling-flash-2.0 (this model)$0.710
  3. gpt-oss-120b (high)$0.900
  4. Llama 4 Scout$0.830
  5. Llama 3.3 70B$1.87
  6. DeepSeek V4 Pro (max)$1.31
  7. MiniMax-M3$1.56
  8. Qwen3 Coder Next$1.90
  9. Nemotron 3 Ultra$3.60
  10. Grok 4.3 (high)$3.95
  11. Kimi K2.6$5.11
  12. GLM-5.2 (max)$6.06
  13. Gemini 3.5 Flash$10.65
  14. GPT-5.5 (xhigh)$35.50
  15. Claude Fable 5 (with fallback)$61.00

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Time per Intelligence Index Task

Weighted average decode time (minutes) per task; excludes TTFT and overhead time · Lower is better · Evaluation results measured independently by Artificial Analysis

15 models in the published order. Ling-flash-2.0 is not marked in this published set of 15 models.
  1. Llama 4 Scout0.198
  2. Llama 3.3 70B0.321
  3. Ling 2.6 Flash1.1
  4. Devstral 21.3
  5. Nemotron 3 Ultra1.4
  6. Grok 4.3 (high)1.6
  7. gpt-oss-120b (high)1.9
  8. Gemini 3.5 Flash2.2
  9. Qwen3 Coder Next2.6
  10. GLM-5.2 (max)2.8
  11. GPT-5.5 (xhigh)3
  12. MiniMax-M33.9
  13. Claude Fable 5 (with fallback)4.8
  14. DeepSeek V4 Pro (max)6.8
  15. Kimi K2.68.2

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Model Size: Total and Active Parameters

Comparison between total model parameters and parameters active during inference · Evaluation results measured independently by Artificial Analysis

12 models; segments: Passive Params, Active Params. Ling-flash-2.0: 103B, row 10 of 12 in the published order.
  • Passive Params
  • Active Params
  1. DeepSeek V4 Pro (max)1,600B
  2. Kimi K2.61,000B
  3. GLM-5.2 (max)753B
  4. Nemotron 3 Ultra550B
  5. MiniMax-M3428B
  6. Devstral 2125B
  7. gpt-oss-120b (high)117B
  8. Llama 4 Scout109B
  9. Ling 2.6 Flash107B
  10. Ling-flash-2.0 (this model)103B
  11. Qwen3 Coder Next79.7B
  12. Llama 3.3 70B70B

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Architecture scale
Larger models generally perform better but cost more to serve. Active parameter count matters for inference efficiency.

Specifications

Architecture and limits

Ling-flash-2.0 technical specifications
SpecificationValueSource
Model typeOpen weightsfaq
ReasoningNofaq
Input modalitiesLing-flash-2.0 supports text input.faq
Output modalitiesLing-flash-2.0 supports text output.faq
Context window130k tokensfaq
Open weights / sourceYes, Ling-flash-2.0 is open weights. The model weights are publicly available and can be downloaded for self-hosting.faq
ParametersLing-flash-2.0 has 103 billion parameters (6.1 billion active).faq
Active parametersLing-flash-2.0 is a Mixture of Experts (MoE) model with 103 billion total parameters, but only 6.1 billion active parameters are used during inference.faq
LicenseLing-flash-2.0 is released under the MIT license. This license allows commercial use.faq
API availabilityYes, Ling-flash-2.0 is available via API through 1 provider.faq

Method

Methodology & Provenance

This page is rendered from the normalized profile and page JSON for Ling-flash-2.0.

Benchmark values are preserved as normalized; only layout, disclosure ordering, and typography are adjusted for readability.

Questions

Model FAQ and technical disclosures

Ling-flash-2.0 was released on September 17, 2025.

Ling-flash-2.0 was created by InclusionAI.

Ling-flash-2.0 scores 10 (estimated) on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 7).

Ling-flash-2.0 generates output at 65.9 tokens per second (based on the median across providers serving the model), which is below average compared to other open weight non-reasoning models of similar size (median: 81.5 t/s).

Ling-flash-2.0 has a time to first token (TTFT) of 2.23s (based on the median across providers serving the model), which is somewhat higher than average compared to other open weight non-reasoning models of similar size (median: 1.59s).

Ling-flash-2.0 costs $0.14 per 1M input tokens (very competitive, median: $0.53) and $0.57 per 1M output tokens (very competitive, median: $1.05), based on the median across providers serving the model.

Ling-flash-2.0 costs $0.14 per 1M input tokens and $0.57 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.18 per 1M tokens. Pricing may vary by provider.

No, Ling-flash-2.0 is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.

Ling-flash-2.0 supports text input.

Ling-flash-2.0 supports text output.

No, Ling-flash-2.0 does not support image input. It can only process text.

No, Ling-flash-2.0 is not multimodal. It only supports text input.

Ling-flash-2.0 has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Ling-flash-2.0 is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Ling-flash-2.0 has 103 billion parameters (6.1 billion active).

Ling-flash-2.0 is a Mixture of Experts (MoE) model with 103 billion total parameters, but only 6.1 billion active parameters are used during inference.

Ling-flash-2.0 is released under the MIT license. This license allows commercial use.

Ling-flash-2.0 achieves a score of 10 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Ling-flash-2.0 is available via API through 1 provider.

Ling-flash-2.0 is available through 1 API provider.

Models