Data as of October 20, 2026 · How the score is built
Released Oct 15, 2026262K contextInternScience Sage-A1.5 model card
Sage-A1.5
Decision readingSage-A1.5 scores 63 out of 100 and ranks #44 of 228. This profile shows 12 source-displayable benchmark rows; its strongest eligible category is Instruction Following at #22. Published weights can be self-hosted, but infrastructure cost varies and is not a comparable API token rate.
Released Oct 15, 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
63/100
field median 49.9#44 of 228 ranked models
Public
#44of 228
Verified —
Price
Self-hosted; infrastructure cost varies
input median $0.72No comparable first-party hosted token rate
Speed
Not measured
field median 93.5 tok/sTime to first token not measured
Context
262Ktokens
field median 256,000Maximum output length is tracked separately
Strongest published evidence
Instruction Following ranks #22. A well-rounded choice across a range of tasks.
Validate before choosing
12 published rows leave some tracked benchmark slots empty. No comparable first-party API token rate is published.
Source-linked · 12 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 |
|---|---|---|---|---|---|---|
| AgenticRank #36 of 132Percentile 73rdWeight 22%6 benchmarksVerified | 57.7 | 6 benchmarks | Verified | |||
| CodingRank #34 of 153Percentile 78thWeight 20%3 benchmarksVerified | 54.0 | 3 benchmarks | Verified | |||
| ReasoningRank Not rankedWeight 17%1 benchmarkVerified | 80.6 | 1 benchmark | Verified | |||
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| KnowledgeWeight 12%1 benchmarkVerified | Score pending | 1 benchmark | Verified | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingRank #22 of 152Percentile 86thWeight 5%1 benchmarkVerified | 91.9 | 1 benchmark | Verified | |||
| MathWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured |
12 of 728 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.
- Agentic6/6 verified
- Coding3/3 verified
- Reasoning1/1 verified
- MultimodalNot measured
- Knowledge1/1 verified
- MultilingualNot measured
- Inst. Following1/1 verified
- MathNot measured
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
Sage-A1.5 category percentile values
- Agentic73rd percentile
- Coding78th percentile
- ReasoningNot eligible
- MultimodalNot eligible
- KnowledgeNot eligible
- MultilingualNot eligible
- Instruction following86th percentile
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#36/132
- Coding#34/153
- ReasoningNot ranked
- MultimodalNot ranked
- KnowledgeNot ranked
- MultilingualNot ranked
- Inst. Following#22/152
- MathNot ranked
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.
Coding3 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench Pro | Score61.2% | Versus best verified row Best verified: Claude Opus 5.5 · 89.9% | Gap28.7 behind | Weight26% ref. weight | Provider exact |
| SciCodeScientific Code Benchmark | Score44.3% | Versus best verified row Best verified: Sakana Fugu · 60.1% | Gap15.8 behind | Weight10% ref. weight | Provider exact |
| SWE Multilingual | Score72.7% | Versus best verified row Best verified: Claude Opus 5.5 · 93.9% | Gap21.2 behind | Weight5% ref. weight | Provider exact |
Agentic6 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| BrowseComp | Score81.7% | Versus best verified row Best verified: Atria Dawn Preview · 92.5% | Gap10.8 behind | Weight8% ref. weight | Provider exact |
| HLE w/ toolsHumanity's Last Exam with tools | Score55.1% | Versus best verified row Best verified: Claude Opus 5.5 · 67.7% | Gap12.6 behind | Weight3% ref. weight | Provider exact |
| GAIAGeneral AI Assistants | Score95.1% | Versus best verified row Best verified: Sage-A1.5 · 95.1% | GapBest verified | WeightDisplay only | Provider exact |
| MLE-Bench Lite | Score65.2% | Versus best verified row Best verified: Atria Dawn Preview · 86.2% | Gap21 behind | WeightDisplay only | Provider exact |
| VITA-Bench | Score45.4% | Versus best verified row Best verified: Qwen3.7 Max · 47.9% | Gap2.5 behind | WeightDisplay only | Provider exact |
| PinchBench | Score90.0% | Versus best verified row Best verified: Pokee-Isaac 28B · 95.7% | Gap5.7 behind | WeightDisplay only | Provider exact |
Reasoning1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| CritPtCritical Physics Tasks | Score20.0% | Versus best verified row Best verified: Muse Wonder Full · 32.9% | Gap12.9 behind | WeightFeeds overall | Provider exact |
Knowledge1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| FrontierScience Research | Score48.3% | Versus best verified row Best verified: Apodex 1.1 · 63.3% | Gap15 behind | WeightDisplay only | Provider exact |
Inst. Following1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| IFBenchInstruction Following Benchmark | Score81.6% | Versus best verified row Best verified: Flint · 89.2% | Gap7.6 behind | WeightWeighted 70% | Provider exact |
Bars run 0–100; the dark tick marks the best source-verified value
All 12 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 publishedInternScience Sage-A1.5 model card
- Context window
- 262KInternScience Sage-A1.5 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
- InternScience publishes BF16 weights under Apache 2.0 on Hugging Face and ModelScope, with FP8 and GGUF editions. The 35B mixture-of-experts model takes text and images, supports a 262,144-token context (512K with YaRN), and offers an opt-in verification mode and a spawn_subagent orchestration tool. The card does not state active parameters.
- 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 recordInternScience Sage-A1.5 model card
- Self-host
- Sage-A1.5 needs ~24GB VRAM (1× NVIDIA RTX 4090 (24GB)).
- 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.
Sage-A1.5 ranks #44 of 228 on the public leaderboard with a score of 62.95/100. It does not yet have enough sourced coverage for a verified position.
Sage-A1.5 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.
InternScience publishes BF16 weights under Apache 2.0 on Hugging Face and ModelScope, with FP8 and GGUF editions. The 35B mixture-of-experts model takes text and images, supports a 262,144-token context (512K with YaRN), and offers an opt-in verification mode and a spawn_subagent orchestration tool. The card does not state active parameters.
Sage-A1.5 sits in the Sage-A1.5 family with Sage-A1.5-F16-GGUF, Sage-A1.5-FP8, Sage-A1.5-Q4_K_M-GGUF, Sage-A1.5-Q8_0-GGUF. 12 of 728 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Instruction Following at #22, while its lowest eligible position is Agentic at #36. a well-rounded choice across a range of tasks.
Last updated October 20, 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.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Questions
How does Sage-A1.5 perform overall in AI benchmarks?
Sage-A1.5 ranks #44 out of 228 models on the public BenchAlign leaderboard, with a score of 62.95/100. Its evidence status is Estimated, and this profile shows 12 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Sage-A1.5 good for knowledge and understanding?
Sage-A1.5 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 Sage-A1.5 good for coding and programming?
Sage-A1.5 ranks #34 out of 153 eligible models for coding and programming, with a public category score of 54/100. 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 Sage-A1.5 good for reasoning and logic?
Sage-A1.5 has source-displayable benchmark coverage for reasoning and logic, 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 Sage-A1.5 good for agentic tool use and computer tasks?
Sage-A1.5 ranks #36 out of 132 eligible models for agentic tool use and computer tasks, with a public category score of 57.7/100. 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 Sage-A1.5 good for instruction following?
Sage-A1.5 ranks #22 out of 152 eligible models for instruction following, with a public category score of 91.9/100. 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 Sage-A1.5 open source?
Sage-A1.5 is an open-weight model from InternScience. 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 Sage-A1.5?
Sage-A1.5 belongs to the Sage-A1.5 family. Related tracked variants include Sage-A1.5-F16-GGUF, Sage-A1.5-FP8, Sage-A1.5-Q4_K_M-GGUF, plus 1 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 Sage-A1.5 have full benchmark coverage on BenchLM?
No. Sage-A1.5 currently has 13 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 Sage-A1.5?
Sage-A1.5 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.