Skip to main content
BenchLM
Data

Data as of November 19, 2026 · How the score is built

Agents-A1

CurrentOpen WeightReasoning

Decision readingAgents-A1 is tracked, but not publicly ranked yet. The profile exposes 6 sourced benchmark rows and leaves unsupported fields blank until a published record exists.

Email me when its price, rank or availability changes.

Capability

Unranked

field median 49Not eligible for a public rank

Price

Not listed

input median $0.60No comparable first-party hosted token rate

Speed

Not measured

field median 93 tok/sTime to first token not measured

Context

262Ktokens

field median 256,000Reported for this model; direct source link not stored

Decision snapshot

How much published evidence this model has, where that evidence is strongest, and what to validate before choosing it. It has no public rank, so no field position is shown.

Strongest published evidence

Published rows are visible, but no category has enough eligible evidence for a comparative rank.

Validate before choosing

6 published rows leave some tracked benchmark slots empty. No comparable first-party API token rate is published.

Source-linked · 6 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 scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank Not rankedWeight 22%3 benchmarksVerified
60.6
CodingWeight 20%0 benchmarksNot measured
Not measured
ReasoningRank Not rankedWeight 17%1 benchmarkVerified
56.2
MultimodalWeight 12%0 benchmarksNot measured
Not measured
KnowledgeRank Not rankedWeight 12%1 benchmarkVerified
68.5
MultilingualWeight 7%0 benchmarksNot measured
Not measured
Inst. FollowingWeight 5%1 benchmarkVerified
Score pending
MathWeight 5%0 benchmarksNot measured
Not measured

6 of 923 tracked benchmark slots have displayable evidence · bars run 0–100

Coverage details

How 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.

  1. Agentic3/3 verified
  2. CodingNot measured
  3. Reasoning1/1 verified
  4. MultimodalNot measured
  5. Knowledge1/1 verified
  6. MultilingualNot measured
  7. Inst. Following1/1 verified
  8. MathNot measured
Verified sourceProvisionalNot measured

Benchmark ledger

Every result says who ran the test. For this model, InternScience ran all 6. Agentic opens by default. The tick marks the best source-verified result in the catalog; provisional leaders do not set the reference.

Overall rank
Not ranked yet
No benchmark-run or independent results yet
Reported by InternScience
6
Shown with that label
Run by a benchmark
0
None yet
Independent runs
0
None yet
Agentic3 rows
Agentic benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
BrowseComp75.5%

Best verified: Atria Dawn Preview · 92.5%

Gap17 behindReported by InternScienceWeighted benchmark9% ref. weight
HLE w/ toolsHumanity's Last Exam with tools47.6%

Best verified: Kimi K3.1 Pro · 70.0%

Gap22.4 behindReported by InternScienceWeighted benchmark3% ref. weight
VITA-Bench38.8%

Best verified: Qwen3.7 Max · 47.9%

Gap9.1 behindReported by InternScienceShown only
Reasoning1 row
Reasoning benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
LongBench v260.2%

Best verified: Inkling 2 Small · 70.6%

Gap10.4 behindReported by InternScienceWeighted benchmark5% ref. weight
Knowledge1 row
Knowledge benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
HLEHumanity's Last Exam47.6%

Best verified: Claude Fable 5.2 · 74.2%

Gap26.6 behindReported by InternScienceWeighted benchmark44% ref. weight
Inst. Following1 row
Inst. Following benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
IFEvalInstruction-Following Eval94.8%

Best verified: LFM2.5-24B-A2B · 96.8%

Gap1.9 behindReported by InternScienceShown only

Outlined bars are lab-reported. Filled bars are benchmark-run or independent. Bars run 0–100; the dark tick marks the best source-verified value.

Lineage

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.

  1. Jun 26, 2026 · you are here

    Agents-A1

    Not publicly ranked · Price not listed

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 published
Context window
262K
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
Not sourced yet
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 record
Self-host
Agents-A1 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.

We track Agents-A1, but the public leaderboard excludes this profile until enough non-generated benchmark coverage is available. Only published rows appear above.

Agents-A1 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.

Official exact-value snapshot from the InternScience Agents-A1 Hugging Face model card and technical report. BenchLM maps only directly comparable public rows: BrowseComp, SciCode, HLE with tools, LongBench-v2, IFEval, tau2-Bench, and VitaBench. FrontierScience-Olympiad, FrontierScience-Research, Seal0, HiPhO, XBench-DS-2510, GAIA, IFBench, MatTools, and MolBench-bind remain outside the current weighted schema.

Agents-A1 sits in the Agents-A1 family with Agents-A1-4B, Agents-A1-F16-GGUF, Agents-A1-FP8, Agents-A1-Q4_K_M-GGUF, Agents-A1-Q8_0-GGUF. 6 of 923 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.

Last updated November 19, 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.

Agents-A1
API / moNo API price on file
Self-host / mo$298
Break-even—
Model the full break-even

Questions

How does Agents-A1 perform overall in AI benchmarks?

Agents-A1 has 6 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 Agents-A1 good for knowledge and understanding?

Agents-A1 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 Agents-A1 good for reasoning and logic?

Agents-A1 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 Agents-A1 good for agentic tool use and computer tasks?

Agents-A1 has source-displayable benchmark coverage for agentic tool use and computer tasks, 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 Agents-A1 good for instruction following?

Agents-A1 has source-displayable benchmark coverage for instruction following, 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 Agents-A1 open source?

Agents-A1 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 Agents-A1?

Agents-A1 belongs to the Agents-A1 family. Related tracked variants include Agents-A1-4B, Agents-A1-F16-GGUF, Agents-A1-FP8, 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 Agents-A1 have full benchmark coverage on BenchLM?

No. Agents-A1 currently has 6 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 Agents-A1?

Agents-A1 has a reported context window of 262K in the exact-model catalog record. The value stays visible, but the profile marks its source link as unavailable instead of presenting it as directly documented. Maximum output length remains separate because providers often publish a different limit.

Compare Agents-A1 with every tracked model1032 comparisons