geekradar
Int
Explore GeekRadar

More to explore

Good gear All topics About us Say hello
GeekRadar / The comparison desk
COMPARE.
AI models
meta

Muse Glimmer 30B

AI model Released

Context window
131,072 tokens
Maximum output
131,072 tokens
Open weights
Yes

Benchmark results

  • Agentic coding SWE-bench Pro · version unreported · reported score · Claude Code · Refined task set with corrected problematic tasks51.20*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Harness
    Claude Code*
    Task set
    Refined task set with corrected problematic tasks*
    Temperature
    1.0*
    Top p
    0.95*
    Context length
    256K tokens*
  • Agentic terminal coding Terminal Bench 2.1 (Terminus) · 2.1 · reported score · Terminus51.70*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Harness
    Terminus*
  • Computer use OSWorld-Verified · version unreported · reported score65.90*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
  • Document intelligence OmniDocBench 1.5 · 1.5 · reported score75.80*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
  • Instruction following IFBench · version unreported · reported score77*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
  • Multidisciplinary reasoning HLE · version unreported · reported score22*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Judge
    GPT-4o*
  • Scientific chart analysis CharXiv (RQ) · version unreported · reported score · Not specified78.80*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Annotation set
    Corrected ground-truth annotations*
    Tool access
    Not specified*
  • Scientific reasoning GPQA Diamond · version unreported · reported score83.50*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*

Run it locally

Inference · memory estimate

Muse Glimmer 30B

Not enough verified architecture information for a reliable VRAM estimate.

API cost

USD / month

Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.

Muse Glimmer 30B

Capabilities

Context window
131,072 tokens
Maximum output
131,072 tokens
Reasoning
Yes
Tool calling
Yes
Open weights
Yes
Knowledge cutoff
2026-01-04
Model license
Apache 2.0

Input modalities

  • text
  • image

Output modalities

  • text

System configuration

Detailed specifications

Specifications · 8 fields
Scientific chart analysis CharXiv (RQ)
suite
Scientific chart analysis CharXiv (RQ)
workload
reported score · Not specified
score
78.8*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Annotation set
Corrected ground-truth annotations*
Tool access
Not specified*
Instruction following IFBench
suite
Instruction following IFBench
workload
reported score
score
77.0*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Document intelligence OmniDocBench 1.5
suite
Document intelligence OmniDocBench 1.5
version
1.5
workload
reported score
score
75.8*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Scientific reasoning GPQA Diamond
suite
Scientific reasoning GPQA Diamond
workload
reported score
score
83.5*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Multidisciplinary reasoning HLE
suite
Multidisciplinary reasoning HLE
workload
reported score
score
22.0*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Judge
GPT-4o*
Computer use OSWorld-Verified
suite
Computer use OSWorld-Verified
workload
reported score
score
65.9*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Agentic terminal coding Terminal Bench 2.1 (Terminus)
suite
Agentic terminal coding Terminal Bench 2.1 (Terminus)
version
2.1
workload
reported score · Terminus
score
51.7*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Harness
Terminus*
Agentic coding SWE-bench Pro
suite
Agentic coding SWE-bench Pro
workload
reported score · Claude Code · Refined task set with corrected problematic tasks
score
51.2*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Harness
Claude Code*
Task set
Refined task set with corrected problematic tasks*
Temperature
1.0*
Top p
0.95*
Context length
256K tokens*

* Provisional data