Skip to main content

Model profile

Soofi S 30B-A3B

Data verified
Overall Score
Unranked
Arena Elo
Not listed
Categories Ranked
0of 8
Price (1M tokens)
$0 in / $0 out
API pricing
Speed
Not listed
Context
1M

Evidence coverage

8 of 299 tracked benchmarks are published. 8 are verified and 0 provisional. 4 of 8 categories are measured.

Updated July 15, 2026Methodology
Published / tracked
8 / 299
Verified
8
Provisional
0
Categories measured
4 / 8

Evidence by category

  • Agentic0 benchmarks
    Not measured
  • Coding1 benchmark
    Verified
  • Reasoning2 benchmarks
    Verified
  • Knowledge4 benchmarks
    Verified
  • Math1 benchmark
    Verified
  • Multilingual0 benchmarks
    Not measured
  • Multimodal0 benchmarks
    Not measured
  • Inst. Following0 benchmarks
    Not measured
Open WeightSelf-hostNon-Reasoning
Confidence:
Low
base

BenchLM is tracking Soofi S 30B-A3B, but this profile is currently excluded from the public leaderboard because it still lacks enough non-generated benchmark coverage to rank safely. Only non-generated public benchmark rows appear below.

Soofi S 30B-A3B is a open weight model with a 1M token context window. It processes queries without explicit chain-of-thought reasoning, offering faster response times and lower token usage.

This profile currently has 8 of 299 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.

Peer position

Exact provisional scores and ranks for the closest listed peers.

Range 45.046.0

  1. Gemini 3.1 Flash-Lite
    Google
    #6146.0
    Gemini 3.1 Flash-Lite is #61 with a score of 46.0.
    Compare
  2. DeepSeek V4 Pro Base
    DeepSeek
    Unranked46.0
    DeepSeek V4 Pro Base is Unranked with a score of 46.0.
    Compare
  3. Soofi S 30B-A3BCurrent model
    Soofi Project
    Unranked45.0
    Soofi S 30B-A3B is Unranked with a score of 45.0.
  4. Laguna XS.2
    Poolside
    Unranked45.0
    Laguna XS.2 is Unranked with a score of 45.0.
    Compare
  5. Mellum2-12B-A2.5B-Instruct
    JetBrains
    Unranked45.0
    Mellum2-12B-A2.5B-Instruct is Unranked with a score of 45.0.
    Compare
  6. Nemotron 3 Nano Omni 30B A3B
    NVIDIA
    Unranked45.0
    Nemotron 3 Nano Omni 30B A3B is Unranked with a score of 45.0.
    Compare
  7. Ternary Bonsai 1.7B
    Prism ML
    Unranked45.0
    Ternary Bonsai 1.7B is Unranked with a score of 45.0.
    Compare

Category percentile

More

Relative position among models eligible for each sourced category. A higher percentile means a stronger position within that category's ranked cohort; 100 is highest.

No eligible category percentile is available from the published evidence yet.

Category evidence

Scores and ranks appear only where this model has published benchmark evidence. Categories without displayable source records remain not measured.

Category scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticWeight 22%0 benchmarksNot measuredNot measured
CodingRank Not rankedWeight 20%1 benchmarkVerified0.0
ReasoningRank Not rankedWeight 17%2 benchmarksVerified0.0
KnowledgeRank Not rankedWeight 12%4 benchmarksVerified35.6
MathRank Not rankedWeight 5%1 benchmarkVerified0.0
MultilingualWeight 7%0 benchmarksNot measuredNot measured
MultimodalWeight 12%0 benchmarksNot measuredNot measured
Inst. FollowingWeight 5%0 benchmarksNot measuredNot measured

Benchmark Details

Rows below have a displayable published verification record. Each source link and provenance note remains in the page HTML while its category is closed. Source-unverified manual rows and generated rows stay hidden.

Coding1 benchmark
HumanEvalProvider exact

Evaluating Large Language Models Trained on Code

73.8%Display only
Source: Soofi Project: Soofi S pretraining reportProvenance: Tables 4 and 5 report HumanEval pass@1 at 73.8 for the selected iter_1056000 Soofi S base checkpoint.
Reasoning2 benchmarks
BBHProvider exact

BIG-Bench Hard

78.8%Display only
Source: Soofi Project: Soofi S pretraining reportProvenance: Tables 4 and 5 report BBH with chain-of-thought at 78.8 for the selected iter_1056000 Soofi S base checkpoint.
DROPProvider exact

Discrete Reasoning Over Paragraphs

66.5%Display only
Source: Soofi Project: Soofi S pretraining reportProvenance: Tables 4 and 5 report DROP exact match at 66.5 for the selected iter_1056000 Soofi S base checkpoint.
Knowledge4 benchmarks
MMLU-ProProvider exact

Massive Multitask Language Understanding Professional

51.4%Weighted 22%
Source: Soofi Project: Soofi S pretraining reportProvenance: Tables 4 and 5 report MMLU-Pro at 51.4 for the selected iter_1056000 Soofi S base checkpoint.
GPQAProvider exact

Graduate-Level Google-Proof Q&A

43.4%Weighted 5%
Source: Soofi Project: Soofi S pretraining reportProvenance: Tables 4 and 5 report GPQA-Diamond at 43.4 for the selected iter_1056000 Soofi S base checkpoint. BenchLM maps it to both GPQA and GPQA-Diamond for comparison consistency.
GPQA-DProvider exact

GPQA Diamond

43.4%Display only
Source: Soofi Project: Soofi S pretraining reportProvenance: Tables 4 and 5 report GPQA-Diamond at 43.4 for the selected iter_1056000 Soofi S base checkpoint.
AGIEvalProvider exact
66.9%Display only
Source: Soofi Project: Soofi S pretraining reportProvenance: Tables 4 and 5 report AGIEval at 66.9 for the selected iter_1056000 Soofi S base checkpoint.
Math1 benchmark
GSM8KProvider exact

Grade School Math 8K

86.1%Display only
Source: Soofi Project: Soofi S pretraining reportProvenance: Tables 4 and 5 report GSM8K at 86.1 for the selected iter_1056000 Soofi S base checkpoint.

Frequently Asked Questions

How does Soofi S 30B-A3B perform overall in AI benchmarks?

Soofi S 30B-A3B has 8 published benchmark scores on BenchLM, but it does not yet have enough non-generated coverage to receive a global overall rank.

Is Soofi S 30B-A3B good for knowledge and understanding?

Soofi S 30B-A3B has visible benchmark coverage in knowledge and understanding, but BenchLM does not currently assign it a global category rank there.

Is Soofi S 30B-A3B good for coding and programming?

Soofi S 30B-A3B has visible benchmark coverage in coding and programming, but BenchLM does not currently assign it a global category rank there.

Is Soofi S 30B-A3B good for mathematics?

Soofi S 30B-A3B has visible benchmark coverage in mathematics, but BenchLM does not currently assign it a global category rank there.

Is Soofi S 30B-A3B good for reasoning and logic?

Soofi S 30B-A3B has visible benchmark coverage in reasoning and logic, but BenchLM does not currently assign it a global category rank there.

Is Soofi S 30B-A3B open source?

Yes, Soofi S 30B-A3B is an open weight model created by Soofi Project, meaning it can be downloaded and run locally or fine-tuned for specific use cases.

Does Soofi S 30B-A3B have full benchmark coverage on BenchLM?

Not yet. Soofi S 30B-A3B currently has 8 published benchmark scores out of the 299 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.

What is the context window size of Soofi S 30B-A3B?

Soofi S 30B-A3B has a context window of 1M, which determines how much text it can process in a single interaction.

Last updated: July 15, 2026 · Runtime metrics stay blank until BenchLM has a sourced snapshot.

Don't miss the next GPT moment

Which models moved up, what is new, and what it costs. One email each week.

Free. One email per week.