Coding
Like-for-like- Muse Spark
- 67.8
- Qwen3.5 397B
- 66.5
- Weighted basis
- 2 vs 2 rows
- Reading
- Muse Spark leads
Muse Spark vs Qwen3.5 397B
Updated August 21, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Muse Spark has the higher public score estimate, 70.6 versus 57.49, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
8 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Coding work
Code generation, repair, and software-engineering tasks
Muse Spark
Muse Spark leads on the same 2 weighted benchmark rows.
Confidence: limited
Long documents
Prompts that approach the documented context limit
Muse Spark
Muse Spark has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate. Muse Spark has no comparable published API token rate.
Confidence: rate-fallback
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
16 Muse Spark only
8 Shared
30 Qwen3.5 397B only
- Shared results
- 8
- Muse Spark only
- 16
- Qwen3.5 397B only
- 30
- Like-for-like categories
- 2 / 8
2 categories use different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.
Multimodal
Like-for-like- Muse Spark
- 82.5
- Qwen3.5 397B
- 79.6
- Weighted basis
- 2 vs 2 rows
- Reading
- Muse Spark leads
Agentic
Directional only- Muse Spark
- 59.0
- Qwen3.5 397B
- 56.5
- Weighted basis
- 1 vs 2 rows
- Reading
- Directional only
Knowledge
Directional only- Muse Spark
- 50.4
- Qwen3.5 397B
- 56.6
- Weighted basis
- 1 vs 4 rows
- Reading
- Directional only
Reasoning
Not comparable- Muse Spark
- 42.5
- Qwen3.5 397B
- 63.2
- Weighted basis
- 1 vs 1 rows
- Reading
- Not comparable
Math
Not comparable- Muse Spark
- 32.9
- Qwen3.5 397B
- 90.6
- Weighted basis
- 2 vs 2 rows
- Reading
- Not comparable
Multilingual
Not comparable- Muse Spark
- Not measured
- Qwen3.5 397B
- 84.7
- Weighted basis
- 0 vs 1 rows
- Reading
- Not comparable
Instruction following
Not comparable- Muse Spark
- Not measured
- Qwen3.5 397B
- 92.6
- Weighted basis
- 0 vs 1 rows
- Reading
- Not comparable
| Category | Muse Spark | Qwen3.5 397B | Weighted basis | Reading |
|---|---|---|---|---|
| Coding | 67.8 | 66.5 | Like-for-like2 vs 2 rows | Muse Spark leads |
| Multimodal | 82.5 | 79.6 | Like-for-like2 vs 2 rows | Muse Spark leads |
| Agentic | 59.0 | 56.5 | Directional only1 vs 2 rows | Directional only |
| Knowledge | 50.4 | 56.6 | Directional only1 vs 4 rows | Directional only |
| Reasoning | 42.5 | 63.2 | Not comparable1 vs 1 rows | Not comparable |
| Math | 32.9 | 90.6 | Not comparable2 vs 2 rows | Not comparable |
| Multilingual | Not measured | 84.7 | Not comparable0 vs 1 rows | Not comparable |
| Instruction following | Not measured | 92.6 | Not comparable0 vs 1 rows | Not comparable |
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
HLE
Knowledge
Terminal-Bench 2.0
Agentic
CharXiv
Multimodal
SWE-bench Pro
Coding
MMMU-Pro
Multimodal
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- Muse Spark
- API rate not published
- Fits in one request
- Qwen3.5 397B
- $0.0024
- Fits in one request
Muse Spark has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Muse Spark
- API rate not published
- Fits in one request
- Qwen3.5 397B
- $0.0408
- Fits in one request
Muse Spark has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Muse Spark
- API rate not published
- Fits in one request
- Cached-input rate unavailable
- Qwen3.5 397B
- $0.168
- Does not fit in one request
- Cached input priced at the published list-input rate
Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate. Muse Spark has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationMuse SparkQwen3.5 397B
Context window
Maximum documented context; output-token limits may be lower.
Muse Spark
262K
Qwen3.5 397B
128K
API model ID
Muse Spark
Not sourced
Qwen3.5 397B
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Muse Spark
No comparable hosted API rate
Qwen3.5 397B
Not published
Documented inputs
Muse Spark
Not sourced
Qwen3.5 397B
Not sourced
Documented outputs
Muse Spark
Not sourced
Qwen3.5 397B
Not sourced
Provider availability
Muse Spark
Not sourced
Qwen3.5 397B
Not sourced
Reasoning profile
Muse Spark
Reasoning
Qwen3.5 397B
Non-Reasoning
Weight access
Muse Spark
Proprietary
Qwen3.5 397B
Open Weight
License
Muse Spark
Proprietary
Qwen3.5 397B
Open Weight
Release date
Muse Spark
2026-04-08
Qwen3.5 397B
2026-02-16
If you already use one of these models
- Deployment change
- The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
- Quality signal
- Muse Spark has the higher public score estimate, 70.6 versus 57.49, but the 90% score intervals overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- Muse Spark has the larger documented window (262K).
Run the same representative tasks against both endpoints before changing production traffic.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence54 rows
Agentic
Terminal-Bench 2.0
Muse Spark59%SourceQwen3.5 397B52.5%SourceMuse Spark leads this result
τ²-bench results
Muse Spark91.5%SourceQwen3.5 397B—Not directly comparable
DeepSearchQA
Muse Spark74.8%SourceQwen3.5 397B—Not directly comparable
CyberGym
Muse Spark43.5%SourceQwen3.5 397B—Not directly comparable
Claw-Eval
Shared sourceMuse Spark63.8%Qwen3.5 397B56.8%Muse Spark leads this result
BrowseComp
Muse Spark—Qwen3.5 397B62%SourceNot directly comparable
QwenClawBench
Muse Spark—Qwen3.5 397B51.8%SourceNot directly comparable
τ³-bench results
Muse Spark—Qwen3.5 397B68.4%SourceNot directly comparable
VITA-Bench
Muse Spark—Qwen3.5 397B43.7%SourceNot directly comparable
DeepPlanning
Muse Spark—Qwen3.5 397B37.6%SourceNot directly comparable
Toolathlon
Muse Spark—Qwen3.5 397B36.3%SourceNot directly comparable
MCP Atlas
Muse Spark—Qwen3.5 397B46.1%SourceNot directly comparable
MCP-Tasks
Muse Spark—Qwen3.5 397B74.2%SourceNot directly comparable
WideResearch
Muse Spark—Qwen3.5 397B74.0%SourceNot directly comparable
Gert Labs
Muse Spark—Qwen3.5 397B46.76%SourceNot directly comparable
ResearchClawBench
Muse Spark—Qwen3.5 397B14.2%SourceNot directly comparable
Coding
SWE-bench Verified
Muse Spark77.4%SourceQwen3.5 397B76.2%SourceMuse Spark leads this result
SWE-bench Pro
Muse Spark52.4%SourceQwen3.5 397B50.9%SourceMuse Spark leads this result
LiveCodeBench Pro
Muse Spark80.0%SourceQwen3.5 397B—Not directly comparable
Vibe Code Bench
Muse Spark19.67%SourceQwen3.5 397B—Not directly comparable
LiveCodeBench v6
Muse Spark—Qwen3.5 397B83.6%SourceNot directly comparable
Reasoning
Knowledge
GPQA-D
Muse Spark89.5%SourceQwen3.5 397B—Not directly comparable
HLE
Muse Spark50.4%SourceQwen3.5 397B28.7%SourceMuse Spark leads this result
HLE w/o tools
Muse Spark42.8%SourceQwen3.5 397B—Not directly comparable
HealthBench Hard
Muse Spark42.8%SourceQwen3.5 397B—Not directly comparable
MedXpertQA (Text)
Muse Spark52.6%SourceQwen3.5 397B—Not directly comparable
GPQA
Muse Spark—Qwen3.5 397B88.4%SourceNot directly comparable
SuperGPQA
Muse Spark—Qwen3.5 397B70.4%SourceNot directly comparable
MMLU-Pro
Muse Spark—Qwen3.5 397B87.8%SourceNot directly comparable
MMLU-Redux
Muse Spark—Qwen3.5 397B94.9%SourceNot directly comparable
C-Eval
Muse Spark—Qwen3.5 397B93%SourceNot directly comparable
Math
FrontierMath v2 (Tiers 1-3)
Muse Spark39.000%SourceQwen3.5 397B—Not directly comparable
FrontierMath v2 (Tier 4)
Muse Spark14.600%SourceQwen3.5 397B—Not directly comparable
AIME26
Muse Spark—Qwen3.5 397B93.3%SourceNot directly comparable
HMMT Feb 2025
Muse Spark—Qwen3.5 397B94.8%SourceNot directly comparable
HMMT Nov 2025
Muse Spark—Qwen3.5 397B92.7%SourceNot directly comparable
HMMT Feb 2026
Muse Spark—Qwen3.5 397B87.9%SourceNot directly comparable
MMAnswerBench
Muse Spark—Qwen3.5 397B80.9%SourceNot directly comparable
Multilingual
Multimodal
CharXiv
Muse Spark86.4%SourceQwen3.5 397B80.8%SourceMuse Spark leads this result
MMMU-Pro
Muse Spark80.4%SourceQwen3.5 397B79%SourceMuse Spark leads this result
ERQA
Muse Spark64.7%SourceQwen3.5 397B—Not directly comparable
SimpleVQA
Muse Spark71.3%SourceQwen3.5 397B—Not directly comparable
ScreenSpot Pro
Muse Spark84.1%SourceQwen3.5 397B65.6%SourceMuse Spark leads this result
ZeroBench
Muse Spark33.0%SourceQwen3.5 397B—Not directly comparable
MedXpertQA (MM)
Muse Spark78.4%SourceQwen3.5 397B—Not directly comparable
MathVision
Muse Spark—Qwen3.5 397B88.6%SourceNot directly comparable
VideoMMMU
Muse Spark—Qwen3.5 397B84.7%SourceNot directly comparable
V*
Muse Spark—Qwen3.5 397B95.8%SourceNot directly comparable
Instruction following
IFEval
Muse Spark—Qwen3.5 397B92.6%SourceNot directly comparable
Frequently asked questions
Which is better, Muse Spark or Qwen3.5 397B?
Muse Spark has the higher public score estimate, 70.6 versus 57.49, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, Muse Spark or Qwen3.5 397B?
Muse Spark leads the like-for-like coding comparison across 2 shared weighted benchmark rows.
Which is better for agentic tasks, Muse Spark or Qwen3.5 397B?
The current agentic tasks averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.
Which costs less, Muse Spark or Qwen3.5 397B?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, Muse Spark or Qwen3.5 397B?
Muse Spark has the larger documented context window: 262K, compared with 128K.
Related comparisons
Last updated August 21, 2026
Watch Muse Spark vs Qwen3.5 397B
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.