Agentic work
Tool use, computer use, and multi-step task completion
Muse Spark
Muse Spark leads on the public agentic lane, 50 to 19.7, with Supported evidence for both models and non-overlapping 90% intervals.
Confidence: stronger
Mistral Medium 3.5 128B vs Muse Spark
Updated September 29, 2026. Rank says Muse Spark is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.
Both of these models will change. Get the price, version and retirement notices for the pair, each with its source. Follow model changes
Share or export
Decision reading
Muse Spark has the higher public score, 60.57 versus 36.09, and the 90% score intervals do not overlap. 4 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 4
- Mistral Medium 3.5 128B only
- 3
- Muse Spark only
- 23
- Like-for-like categories
- 2 / 8
Estimated: Mistral Medium 3.5 128B · Supported: Muse SparkHow the comparison works
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.
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
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Mistral Medium 3.5 128B is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limitedChat 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-ratesCache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: rate-fallbackRepository 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
Which one for a specific job
Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.
The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.
26.6Mistral Medium 3.5 128B53.3Muse Spark
Directional only · BenchAlign v5.7
Muse Spark scores higher, but at least one score rests on Estimated evidence or a different benchmark set. Directional only, no winner.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Same basis rules as the category table below
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
3 Mistral Medium 3.5 128B only
4 Shared
23 Muse Spark only
2 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
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.
MMLU-Pro (Vals)Knowledge
Normalized gap 12.0Mistral Medium 3.5 128B:75.3%Muse Spark:87.3%SWE-bench VerifiedCoding
Normalized gap 0.2Mistral Medium 3.5 128B:77.6%Muse Spark:77.4%
Bars run 0–100 on each benchmark’s normalized display scale
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign v5.7 lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.
Agentic
Like-for-like- Mistral Medium 3.5 128B
- 19.7
- Supported · #100/117
- Muse Spark
- 50.0
- Supported · #39/117
- Basis
- BenchAlign v5.7 lane · 3 vs 5 public rows
- Reading
- Muse Spark leads
Knowledge
Like-for-like- Mistral Medium 3.5 128B
- 33.6
- Supported · #121/169
- Muse Spark
- 60.5
- Supported · #40/169
- Basis
- BenchAlign v5.7 lane · 2 vs 7 public rows
- Reading
- Muse Spark leads
Coding
Directional only- Mistral Medium 3.5 128B
- 26.6
- Estimated · #102/143
- Muse Spark
- 53.3
- Supported · #35/143
- Basis
- BenchAlign v5.7 lane · 2 vs 5 public rows
- Reading
- Directional only
Instruction following
Directional only- Mistral Medium 3.5 128B
- 82.6
- #48/124
- Muse Spark
- 91.9
- #8/124
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Directional only
Reasoning
Not comparable- Mistral Medium 3.5 128B
- 69.9
- Unranked · 2 rankable rows
- Muse Spark
- 53.4
- Unranked · 3 rankable rows
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Mistral Medium 3.5 128B
- 56.7
- Unranked · 1 rankable row
- Muse Spark
- 78.5
- #14/50
- Basis
- Provisional lane · 0 vs 2 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Mistral Medium 3.5 128B
- Not ranked
- Muse Spark
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Mistral Medium 3.5 128B
- Not ranked
- Muse Spark
- 55.1
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 0 vs 2 weighted rows
- Reading
- Not comparable
| Category | Mistral Medium 3.5 128B | Muse Spark | Basis | Reading |
|---|---|---|---|---|
| Agentic | 19.7Supported · #100/117 | 50.0Supported · #39/117 | Like-for-likeBenchAlign v5.7 lane · 3 vs 5 public rows | Muse Spark leads |
| Knowledge | 33.6Supported · #121/169 | 60.5Supported · #40/169 | Like-for-likeBenchAlign v5.7 lane · 2 vs 7 public rows | Muse Spark leads |
| Coding | 26.6Estimated · #102/143 | 53.3Supported · #35/143 | Directional onlyBenchAlign v5.7 lane · 2 vs 5 public rows | Directional only |
| Instruction following | 82.6#48/124 | 91.9#8/124 | Directional onlyProvisional lane · 0 vs 0 weighted rows | Directional only |
| Reasoning | 69.9Unranked · 2 rankable rows | 53.4Unranked · 3 rankable rows | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Multimodal | 56.7Unranked · 1 rankable row | 78.5#14/50 | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | Not ranked | 55.1Unranked · 2 rankable rows | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.7) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Supported evidence per lane · bars run 0–100Methodology
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
- Mistral Medium 3.5 128B
- $0.00525
- Fits in one request
- Muse Spark
- API rate not published
- Fits in one request
Muse Spark has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Mistral Medium 3.5 128B
- $0.0975
- Fits in one request
- Muse Spark
- API rate not published
- 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
- Mistral Medium 3.5 128B
- $0.405
- Fits in one request
- Cached input priced at the published list-input rate
- Muse Spark
- API rate not published
- Fits in one request
- Cached-input rate unavailable
Mistral Medium 3.5 128B has no published cached-input rate, so cached tokens use its listed input rate. Muse Spark has no comparable published API token rate.
Cached input falls back to the list input rate only where a cached rate is unpublished
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationMistral Medium 3.5 128BMuse Spark
Context window
Maximum documented context; output-token limits may be lower.
Mistral Medium 3.5 128B
256K
Muse Spark
262K
API model ID
Mistral Medium 3.5 128B
Not sourced
Muse Spark
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Mistral Medium 3.5 128B
Not published
Muse Spark
No comparable hosted API rate
Documented inputs
Mistral Medium 3.5 128B
Not sourced
Muse Spark
Not sourced
Documented outputs
Mistral Medium 3.5 128B
Not sourced
Muse Spark
Not sourced
Provider availability
Mistral Medium 3.5 128B
Not sourced
Muse Spark
Not sourced
Reasoning profile
Mistral Medium 3.5 128B
Reasoning
Muse Spark
Reasoning
Weight access
Mistral Medium 3.5 128B
Open Weight
Muse Spark
Proprietary
License
Mistral Medium 3.5 128B
Open Weight
Muse Spark
Proprietary
Release date
Mistral Medium 3.5 128B
2026-04-29
Muse Spark
2026-04-08
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, 60.57 versus 36.09, and the 90% score intervals do not 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.
Questions
Which is better, Mistral Medium 3.5 128B or Muse Spark?
Muse Spark has the higher public score, 60.57 versus 36.09, and the 90% score intervals do not overlap. That is the clearest overall quality signal in the current public data.
Which is better for coding, Mistral Medium 3.5 128B or Muse Spark?
Muse Spark scores higher for coding on the public lane, 53.3 to 26.6. Mistral Medium 3.5 128B is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which is better for agentic tasks, Mistral Medium 3.5 128B or Muse Spark?
Muse Spark leads the public agentic tasks lane, 50 to 19.7, with Supported evidence for both models and non-overlapping 90% intervals.
Which costs less, Mistral Medium 3.5 128B or Muse Spark?
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, Mistral Medium 3.5 128B or Muse Spark?
Muse Spark has the larger documented context window: 262K, compared with 256K.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence30 rows
Agentic
τ³-bench results
Mistral Medium 3.5 128B91.4%SourceMuse Spark—Not directly comparable
Gert Labs
Mistral Medium 3.5 128B39.10%SourceMuse Spark—Not directly comparable
Terminal-Bench 2.1 (Vals)
Mistral Medium 3.5 128B39.0%SourceMuse Spark—Not directly comparable
Terminal-Bench 2.0
Mistral Medium 3.5 128B—Muse Spark59%SourceNot directly comparable
τ²-bench results
Mistral Medium 3.5 128B—Muse Spark91.5%SourceNot directly comparable
DeepSearchQA
Mistral Medium 3.5 128B—Muse Spark74.8%SourceNot directly comparable
CyberGym
Mistral Medium 3.5 128B—Muse Spark43.5%SourceNot directly comparable
Claw-Eval
Mistral Medium 3.5 128B—Muse Spark63.8%SourceNot directly comparable
Coding
SWE-bench Verified
Mistral Medium 3.5 128B77.6%SourceMuse Spark77.4%SourceMistral Medium 3.5 128B leads this result
SWE-bench (Vals)
Mistral Medium 3.5 128B66.4%SourceMuse Spark74.4%SourceMuse Spark leads this result
SWE-bench Pro
Mistral Medium 3.5 128B—Muse Spark52.4%SourceNot directly comparable
LiveCodeBench Pro
Mistral Medium 3.5 128B—Muse Spark80.0%SourceNot directly comparable
Vibe Code Bench
Mistral Medium 3.5 128B—Muse Spark19.67%SourceNot directly comparable
Reasoning
ARC-AGI-2
Mistral Medium 3.5 128B—Muse Spark42.5%SourceNot directly comparable
Multimodal
CharXiv
Mistral Medium 3.5 128B—Muse Spark86.4%SourceNot directly comparable
MMMU-Pro
Mistral Medium 3.5 128B—Muse Spark80.4%SourceNot directly comparable
ERQA
Mistral Medium 3.5 128B—Muse Spark64.7%SourceNot directly comparable
SimpleVQA
Mistral Medium 3.5 128B—Muse Spark71.3%SourceNot directly comparable
ScreenSpot Pro
Mistral Medium 3.5 128B—Muse Spark84.1%SourceNot directly comparable
ZeroBench
Mistral Medium 3.5 128B—Muse Spark33.0%SourceNot directly comparable
MedXpertQA (MM)
Mistral Medium 3.5 128B—Muse Spark78.4%SourceNot directly comparable
Knowledge
GPQA Diamond (Vals)
Mistral Medium 3.5 128B34.8%SourceMuse Spark89.6%SourceMuse Spark leads this result
MMLU-Pro (Vals)
Mistral Medium 3.5 128B75.3%SourceMuse Spark87.3%SourceMuse Spark leads this result
GPQA-D
Mistral Medium 3.5 128B—Muse Spark89.5%SourceNot directly comparable
HLE
Mistral Medium 3.5 128B—Muse Spark50.4%SourceNot directly comparable
HLE w/o tools
Mistral Medium 3.5 128B—Muse Spark42.8%SourceNot directly comparable
HealthBench Hard
Mistral Medium 3.5 128B—Muse Spark42.8%SourceNot directly comparable
MedXpertQA (Text)
Mistral Medium 3.5 128B—Muse Spark52.6%SourceNot directly comparable
Math
30 public results · 4 shared
Watch Mistral Medium 3.5 128B vs Muse Spark
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.
Last updated September 29, 2026