Data as of October 29, 2026 · How the score is built
Released Aug 4, 2026262K contextInclusionAI Ling 3.0 Flash FP8 model card
Ling 3.0 Flash FP8
Decision readingLing 3.0 Flash FP8 is tracked, but not publicly ranked yet. The profile exposes 4 sourced benchmark rows and leaves unsupported fields blank until a published record exists.
Released Aug 4, 2026 — see all recent releases
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Capability
Unranked
field median 50.5Not eligible for a public rank
Price
API rate not published
input median $0.75No comparable first-party hosted token rate
Speed
Not measured
field median 92 tok/sTime to first token not measured
Context
262Ktokens
field median 256,000Maximum output length is tracked separately
Strongest published evidence
Instruction Following ranks #75. A well-rounded choice across a range of tasks.
Validate before choosing
4 published rows leave some tracked benchmark slots empty. No comparable first-party API token rate is published.
Source-linked · 4 displayable benchmark rows
Category score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticWeight 22%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| CodingRank Not rankedWeight 20%1 benchmarkVerified | 41.4 | 1 benchmark | Verified | |||
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| KnowledgeRank Not rankedWeight 12%2 benchmarksVerified | 73.6 | 2 benchmarks | Verified | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingRank #75 of 157Percentile 53rdWeight 5%1 benchmarkVerified | 72.6 | 1 benchmark | Verified | |||
| MathWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured |
4 of 782 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- AgenticNot measured
- Coding1/1 verified
- ReasoningNot measured
- MultimodalNot measured
- Knowledge2/2 verified
- MultilingualNot measured
- Inst. Following1/1 verified
- MathNot measured
Benchmark ledger
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Coding1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SciCodeScientific Code Benchmark | Score40.4% | Versus best verified row Best verified: Sakana Fugu · 60.1% | Gap19.7 behind | Weight10% ref. weight | Provider exact |
Knowledge2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| GPQAGraduate-Level Google-Proof Q&A | Score84% | Versus best verified row Best verified: GPT-6 Astra · 96% | Gap12 behind | Weight3% ref. weight | Provider exact |
| GPQA-DGPQA Diamond | Score84.0% | Versus best verified row Best verified: GPT-6 Astra · 96.0% | Gap12 behind | WeightDisplay only | Provider exact |
Inst. Following1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| IFBenchInstruction Following Benchmark | Score73.4% | Versus best verified row Best verified: Flint · 89.2% | Gap15.8 behind | WeightWeighted 70% | Provider exact |
Bars run 0–100; the dark tick marks the best source-verified value
All 4 rowsLineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not publishedInclusionAI Ling 3.0 Flash FP8 model card
- Context window
- 262KInclusionAI Ling 3.0 Flash FP8 model card
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not sourced yet
- Availability
- InclusionAI publishes the blockwise FP8 checkpoint under the MIT License on Hugging Face. The repository documents two-GPU SGLang serving and the same 262,144-token context as the BF16 checkpoint.
- Cloud regions
- Not tracked yet
- Lifecycle
- Current
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing recordInclusionAI Ling 3.0 Flash FP8 model card
- Self-host
- Open weights available; hardware estimate not sourced
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track Ling 3.0 Flash FP8, but the public leaderboard excludes this profile until enough non-generated benchmark coverage is available. Only published rows appear above.
Ling 3.0 Flash FP8 is a open weight model with a 262K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
InclusionAI publishes the blockwise FP8 checkpoint under the MIT License on Hugging Face. The repository documents two-GPU SGLang serving and the same 262,144-token context as the BF16 checkpoint.
Official exact-value quantization snapshot from InclusionAI's Ling-3.0-flash-fp8 Hugging Face model card. The card independently evaluates the blockwise FP8 checkpoint on GPQA-Diamond, IFBench, SciCode, and ArcPrize. BenchLM maps the first three to exact existing keys and leaves ArcPrize outside the schema. The row remains non-ranking because the FP8-specific evidence covers only three supported benchmark families; no BF16 launch-table scores are copied onto the quantized variant.
Ling 3.0 Flash FP8 sits in the Ling 3.0 family with Ling 3.0 Flash, Ling 3.0 Flash VL, Ling 3.0 Flash Sante, Ling 3.0 Tiny, Ling 3.0 Flash Fin. 4 of 782 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Instruction Following at #75. a well-rounded choice across a range of tasks.
Last updated October 29, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.
Estimate VRAM from known parametersQuestions
How does Ling 3.0 Flash FP8 perform overall in AI benchmarks?
Ling 3.0 Flash FP8 has 4 source-displayable benchmark rows, but it does not qualify for a public overall rank. The available rows remain visible by category without being converted into a site-wide score. Missing evidence stays blank instead of being estimated from an earlier model.
Is Ling 3.0 Flash FP8 good for knowledge and understanding?
Ling 3.0 Flash FP8 has source-displayable benchmark coverage for knowledge and understanding, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Ling 3.0 Flash FP8 good for coding and programming?
Ling 3.0 Flash FP8 has source-displayable benchmark coverage for coding and programming, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Ling 3.0 Flash FP8 good for instruction following?
Ling 3.0 Flash FP8 ranks #75 out of 157 eligible models for instruction following, with a public category score of 72.6/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Ling 3.0 Flash FP8 open source?
Ling 3.0 Flash FP8 is an open-weight model from InclusionAI. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Which sibling models are related to Ling 3.0 Flash FP8?
Ling 3.0 Flash FP8 belongs to the Ling 3.0 family. Related tracked variants include Ling 3.0 Flash, Ling 3.0 Flash VL, Ling 3.0 Flash Sante, plus 2 more. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does Ling 3.0 Flash FP8 have full benchmark coverage on BenchLM?
No. Ling 3.0 Flash FP8 currently has 4 source-displayable rows. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Ling 3.0 Flash FP8?
Ling 3.0 Flash FP8 has a documented context window of 262K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.