Meta
·Released on Aug 9, 2026Muse Glimmer 30B API Benchmarks, Pricing & Provider Data
Compare Muse Glimmer 30B with another model
Choose a model to open its comparison page.
LLM
Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-ho...
Quality
#55of 100
57.0
LMSpeed score
Category Performance
Observed-capability estimates with uncertainty reported separately from independent benchmark families.
- Coverage
- 7 / 8
- Methodology
- V3.0
#1Reasoning56.1Estimated2/4 Measured dimensions
80% interval: 45.3–66.9
abstract logicPrior only
scientific causal54
critpt · critpt / gpqa · aa_gpqa_diamond / hle · aa_hle
multistep constraintsPrior only
evidence verification58.2
lcr · lcr
#2Instruction following55.1Provisional1/4 Measured dimensions
80% interval: 39.1–71.1
constraint followingPrior only
structured outputPrior only
novel instruction generalization55.1
ifbench · if_bench
long multiturn instructionPrior only
#3Agents47.7Estimated2/4 Measured dimensions
80% interval: 37.5–57.9
planningPrior only
tool use48.8
mcp_atlas · mcp_atlas / tau · aa_tau3_banking
environment execution46.5
deep_search_qa · deep_search_qa / gdpval_aa · benchlm_agentic_gdpval_aa / osworld · os_world_verified
recovery reliabilityPrior only
#4Coding46.6RatedGlobal rank #344/4 Measured dimensions
80% interval: 39.7–53.5
code generation54.8
scicode · aa_sci_code
repository engineering42
swe_pro · swe_pro
debugging testing49.2
swe_verified · swe_verified
tooling quality40.4
terminalbench · aa_terminal_bench21
#5Multimodal46.4Estimated3/4 Measured dimensions
80% interval: 37.1–55.8
perception ocrPrior only
document spatial46.8
charxiv · charxiv
visual reasoning42.8
mmmu_pro · mmmu_pro
video action49.7
screenspot · screen_spot_pro
#6Math46.1Provisional1/4 Measured dimensions
80% interval: 29.8–62.4
foundational mathPrior only
competition math46.1
aime · aime2026
proof frontierPrior only
applied tool mathPrior only
#7Knowledge43.3Provisional1/4 Measured dimensions
80% interval: 27.3–59.3
broad knowledge43.3
aa_omniscience · aa_omniscience_index
professional knowledgePrior only
factualityPrior only
retrieval open bookPrior only
No data:Multilingual
Specifications
Input and output token limits for this model, plus how it ranks on long-context understanding.
INPUT
131.1Ktokens
≈ 157.3 pages of text
OUTPUT
131.1Ktokens
8K128K1M4M
131.1K
Features
Technical Details
- Input
- Output
- Total parameters
- 30B
- Released
- Aug 2026
- Tokenizer
- Other
- Architecture
- text+image->text
- Moderated
- No
- Supported parameters
- frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Rankings
Excels at
It's decent at
Falls behind in
Detailed scores
Updated: Aug 23, 2026Overall
undefined metric} other undefined metrics}}
Overall score57.0#55 / 100SkillsBench44.3#3 / 3
Overall
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Agents
V3.0undefined metric} other undefined metrics}} · Estimated
Score47.780% interval37.5–57.92/4 Measured dimensionsAgentic score51.7#46 / 69MCP Atlas75.5#11 / 31
Agents
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Planning & decompositionPrior only
Tool use48.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mcp_atlas · mcp_atlas · z 0.21 · q 1.00
tau · aa_tau3_banking · z -1.49 · q 1.00
Environment & long-horizon execution46.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
deep_search_qa · deep_search_qa · z -0.36 · q 1.00
gdpval_aa · benchlm_agentic_gdpval_aa · z -0.64 · q 1.00
osworld · os_world_verified · z -0.62 · q 1.00
Recovery & completion reliabilityPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Coding
V3.0undefined metric} other undefined metrics}} · Rated
Score46.6#3480% interval39.7–53.54/4 Measured dimensionsCoding score47.3#62 / 81SWE-bench Pro51.2#44 / 49
Coding
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Code generation54.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
scicode · aa_sci_code · z 0.50 · q 1.00
Repository engineering42
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
swe_pro · swe_pro · z -1.01 · q 1.00
Debugging & testing49.2
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
swe_verified · swe_verified · z -0.16 · q 1.00
Tool-assisted development & quality40.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
terminalbench · aa_terminal_bench21 · z -1.53 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
AA Terminal-Bench 2.1V3 evidence
LMSpeed rank#16
Score51.7
Sourcebenchlm.ai
UpdatedAug 20, 2026
Confidence2
Reasoning
V3.0undefined metric} other undefined metrics}} · Estimated
Score56.180% interval45.3–66.92/4 Measured dimensionsAA-LCR80.0#5 / 100CritPt2.6#53 / 83
Reasoning
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Abstract logicPrior only
Scientific & causal reasoning54
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
critpt · critpt · z -0.03 · q 1.00
gpqa · aa_gpqa_diamond · z 0.39 · q 1.00
hle · aa_hle · z 0.37 · q 1.00
Multi-step constraintsPrior only
Evidence integration & verification58.2
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
lcr · lcr · z 0.73 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Score43.380% interval27.3–59.31/4 Measured dimensionsArtificial Analysis Intelligence Index35.1#62 / 111AA-GPQA Diamond83.5#64 / 108
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Broad knowledge43.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aa_omniscience · aa_omniscience_index · z -0.98 · q 1.00
Professional knowledgePrior only
FactualityPrior only
Retrieval & open-book usePrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Artificial Analysis Intelligence IndexV3 evidence
LMSpeed rank#62
Score35.1
Sourcebenchlm.ai
UpdatedAug 23, 2026
Confidence2
AA-Omniscience Hallucination Rate
LMSpeed rank#37
Score81.9
Sourcebenchlm.ai
UpdatedAug 23, 2026
Confidence2
Math
V3.0undefined metric} other undefined metrics}} · Provisional
Score46.180% interval29.8–62.41/4 Measured dimensionsMath score75.5#13 / 62AIME2694.7#10 / 14
Math
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Foundational mathPrior only
Competition math46.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aime · aime2026 · z -0.52 · q 1.00
Advanced proofsPrior only
Applied & tool-assisted mathPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
No data80% interval30.8–69.20/4 Measured dimensions
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
Dimensions and evidence
Cross-language understandingPrior only
Multilingual generationPrior only
Reasoning transferPrior only
Low-resource robustnessPrior only
Multimodal
V3.0undefined metric} other undefined metrics}} · Estimated
Score46.480% interval37.1–55.83/4 Measured dimensionsMultimodal Grounded score48.9#37 / 44CharXiv78.8#21 / 29
Multimodal
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Perception & OCRPrior only
Document & spatial understanding46.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
charxiv · charxiv · z -0.30 · q 1.00
Visual reasoning42.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmmu_pro · mmmu_pro · z -1.02 · q 1.00
Video & grounded action49.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
screenspot · screen_spot_pro · z -0.34 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Instruction following
V3.0undefined metric} other undefined metrics}} · Provisional
Score55.180% interval39.1–71.11/4 Measured dimensionsInstruction Following score84.5#17 / 25IFBench77.0#8 / 14
Instruction following
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Constraint followingPrior only
Structured outputPrior only
Novel-instruction generalization55.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifbench · if_bench · z 0.40 · q 1.00
Multi-turn & long instructionsPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
OpenRouter endpoints
5 endpointsThird-party OpenRouter endpoint data, shown separately from LMSpeed measurements. Some 30-minute live performance fields only appear after syncing with an OpenRouter API key.
| Provider endpoint | Input | Output | 1d uptime | 30m latency | 30m throughput | Context / output |
|---|---|---|---|---|---|---|
Parasail parasail/bf16 | $0.300/M | $1.10/M | 100.0% | — | — | undefined tokens / undefined tokens |
DeepInfra deepinfra/bf16 | $0.300/M | $1.20/M | 100.0% | — | — | undefined tokens / undefined tokens |
Phala phala | $0.300/M | $1.10/M | 99.9% | — | — | undefined tokens / undefined tokens |
Together together | $0.350/M | $1.50/M | 99.9% | — | — | undefined tokens / — |
Fireworks fireworks | $0.350/M | $1.50/M | 98.1% | — | — | undefined tokens / — |