Muse Spark 1.3

MetaKey muse-spark-1.3Also reported as meta/muse-spark-1.3, sk/muse-spark-1.3, nutaraline/muse-spark-1.3

Rank
#61
by active users
Active users
21
Down 6 from 27 active users in the previous 30 days. vs previous 30 days
Adoption
2.2%
of 958 accounts
Tokens
6.7B
cache included
Est. cost
$747
at public list prices
First seen
Sep 2, 2026
in any submission

Oct 14 – Nov 12, 2026 (UTC)958 accounts100% of tokens from an identified modelData through Nov 12, 2026, 00:36 UTCHow we count

Daily active users

Distinct accounts that used this model on each UTC day.

0 active users on Nov 12, 2026

Daily active users for Muse Spark 1.3, Oct 14, 2026 to Nov 12, 2026. Peak 7. Use the left and right arrow keys to read each day.840
Show daily values
Daily active users and tokens for Muse Spark 1.3
Date (UTC)Active usersTokens
Oct 14, 20265545.7M
Oct 15, 20264468.4M
Oct 16, 202661.2B
Oct 17, 202662.5B
Oct 18, 20264819K
Oct 19, 202648M
Oct 20, 2026422.7M
Oct 21, 20262633K
Oct 22, 20262402K
Oct 23, 2026540M
Oct 24, 20263873K
Oct 25, 202622.6M
Oct 26, 2026273.9K
Oct 27, 202621.8M
Oct 28, 202600
Oct 29, 20261158K
Oct 30, 2026136.2K
Oct 31, 20263742K
Nov 1, 202618.5K
Nov 2, 202652M
Nov 3, 2026320.1M
Nov 4, 20263542K
Nov 5, 20262190.2M
Nov 6, 2026135.4M
Nov 7, 20267515.2M
Nov 8, 20262711.9M
Nov 9, 20264279.8M
Nov 10, 20263145.3M
Nov 11, 2026118.9M
Nov 12, 202600

Token mix

How the 6.7B tokens split. Cache reads are usually most of an agent's tokens and are priced far below input.

  • Input5.8%
  • Output0.3%
  • Cache read94%
  • Cache write0%
  • Reasoning<0.1%

Agents that carry Muse Spark 1.3

Each agent's share of this model's tokens, and how many accounts used the model through it.

  1. Codebuff55% of this model's tokens1active users
  2. Muse Code30% of this model's tokens15active users
  3. Oh My Pi13% of this model's tokens3active users
  4. Pi1.2% of this model's tokens2active users
  5. OpenCode0.6% of this model's tokens2active users
  6. Reasonix0.3% of this model's tokens1active users
  7. Claude Code<0.1% of this model's tokens1active users
  8. Junie<0.1% of this model's tokens1active users

Benchmarks

The best result for Muse Spark 1.3 in each benchmark Tokscale tracks. Each benchmark has its own unit, run conditions and source, so compare a score only with other models on the same benchmark. Rank is among the models Tokscale matched in that benchmark. Agent rows also show Tokscale usage through that agent, side by side with the score. Usage does not explain the score.

All benchmarks
  • Epoch Capabilities IndexOverall · Epoch AI
    157.2
    Rank #23 of 160 matched modelsNo conditions reportedModel released Sep 2, 2026Source: epoch.ai (opens in a new tab)
  • Terminal-Bench 4.0Agentic coding · Terminal-Bench
    14.5%
    Rank #20 of 23 matched modelsXHigh · Muse CodePublished Oct 6, 2026Source: tbench.ai (opens in a new tab)

    Tokscale usage: Muse Code carried 30% of Muse Spark 1.3 tokens in the last 30 days (15 active users).

  • CursorBenchAgentic coding · Cursor, collected by Epoch AI
    41.6%
    Rank #11 of 18 matched modelsMaxModel released Sep 2, 2026Source: cursor.com (opens in a new tab)
    Show 5 more results with other run conditionsHide the other results
    • Listed as Muse Spark 1.3 Extra High
      37.5%
      XHighModel released Sep 2, 2026Source: cursor.com (opens in a new tab)
    • Listed as Muse Spark 1.3 High
      33.4%
      HighModel released Sep 2, 2026Source: cursor.com (opens in a new tab)
    • Listed as Muse Spark 1.3 Medium
      32.6%
      MediumModel released Sep 2, 2026Source: cursor.com (opens in a new tab)
    • Listed as Muse Spark 1.3 Low
      29.3%
      LowModel released Sep 2, 2026Source: cursor.com (opens in a new tab)
    • Listed as Muse Spark 1.3 Minimal
      24.3%
      MinimalModel released Sep 2, 2026Source: cursor.com (opens in a new tab)
  • FrontierSWEAgentic coding · FrontierSWE, collected by Epoch AI
    25.8%
    Rank #13 of 21 matched modelsMax · proximusModel released Sep 2, 2026Source: frontierswe.com (opens in a new tab)
  • WebDev ArenaWeb development · LMArena
    1,659
    Rank #13 of 110 matched modelsMaxPublished Oct 19, 2026Source: huggingface.co (opens in a new tab)
    Show 1 more result with other run conditionsHide the other results
    • Listed as Muse Spark 1.3 (xHigh)
      1,621
      XHighPublished Oct 19, 2026Source: huggingface.co (opens in a new tab)
  • Agent ArenaAgents · LMArena
    +0.03
    Rank #15 of 48 matched modelsMax · Arena Agent ModePublished Oct 19, 2026Source: huggingface.co (opens in a new tab)
  • FrontierMath Tiers 1–3 (v2)Reasoning · Epoch AI
    74.4%
    Rank #22 of 86 matched modelsXHighRun date Sep 16, 2026Source: epoch.ai (opens in a new tab)

    Near its ceiling: the top score is 93.7%, and Epoch AI estimates the reachable maximum at 95%. Differences between top models are small.

    Show 1 more result with other run conditionsHide the other results
    • Listed as muse-spark-1.3_max
      74.0%
      MaxRun date Sep 18, 2026Source: epoch.ai (opens in a new tab)

Not evaluated: GPQA Diamond, Humanity's Last Exam, SimpleQA Verified, DeepSWE. A missing result is not a score of zero.

Pricing

List prices from models.dev in US dollars per 1M tokens, retrieved Oct 21, 2026. Meta is the lab's own API and the reference price for the estimates below.

Price at real agent usageEstimate at list prices

What 1M tokens cost at the reference list price when they split into input, output and cache the way Tokscale users' agents used them. It is not what anyone paid.

At this model's mix
$0.23
Est. per 1M tokens
94% cache reads
At the community mix
$0.20
Est. per 1M tokens
96% cache reads
3:1 input/output blend
$2
Per 1M tokens at list prices
No cache

Cache reads were 94% of Muse Spark 1.3 tokens in the last 30 days and 96% of all tokens. The 3:1 blend, (3 × input + output) ÷ 4, is the usual comparison and ignores cache.

Assumptions

  • Meta list prices per 1M tokens: input $1.25, output $4.25, cache read $0.15.
  • This model's mix is its own input, output, cache and reasoning tokens in the last 30 days.
  • The community mix is the tokens of every identified model in the last 30 days.
  • This is an estimate at list prices, not what anyone paid.
  • Reasoning tokens are priced as output.
  • Long-context tiers, batch pricing and discounts are not applied.
  • Token mixes come from public Tokscale submissions. Hidden accounts are excluded.
  • No cache-write price is listed, so cache writes are priced as input.

List prices by provider

  • MetaReferencemuse-spark-1.3
    $1.25 input per 1M tokens$4.25 output per 1M tokens$0.15 cache read per 1M tokensCache write not listed1.05M tokens of context
  • OpenRoutermeta/muse-spark-1.3
    $1.25 input per 1M tokens$4.25 output per 1M tokens$0.15 cache read per 1M tokensCache write not listed1.05M tokens of context
  • Vercel AI Gatewaymeta/muse-spark-1.3
    $1.25 input per 1M tokens$4.25 output per 1M tokens$0.15 cache read per 1M tokensCache write not listed1.05M tokens of context

Specs

From models.dev (MIT), retrieved Oct 21, 2026, as the lab or the first listed provider reports them. A dash means models.dev does not list the value.

Context window
1.05M tokens
Max output
131K tokens
Input
Text, Image, Video, PDF, Audio
Output
Text
Knowledge cutoff
— not listed
Release date
Sep 2, 2026
Open weights
No
Reasoning
Yes
Tool calling
Yes

Top public profiles

Public profiles that used Muse Spark 1.3 most in this period, by tokens. Hidden accounts are never listed.

  1. 1Baptiste@BaptisteFaisy3.7Btokens—est. cost5active daysno price · 5 active days
  2. 2Khaliq@khaliqgant1.7Btokens$348est. cost9active days$348 est. cost · 9 active days
  3. 3insterbridge@insterbridge885.4Mtokens$231est. cost4active days$231 est. cost · 4 active days
  4. 4Ziyang Fang@Fangziyang0910249.6Mtokens$54est. cost6active days$54 est. cost · 6 active days
  5. 5AllThatMagic@AllThatMagic81.4Mtokens—est. cost4active daysno price · 4 active days
  6. 6Yags@whyyagswhy68.1Mtokens$61est. cost5active days$61 est. cost · 5 active days
  7. 7Sunghyun Cho@anaclumos37.4Mtokens$28est. cost5active days$28 est. cost · 5 active days
  8. 8Rupreet Gujral@rupreetg36.9Mtokens$8est. cost1active days$8 est. cost · 1 active days
  9. 9Sam Foreman@saforem24.6Mtokens$2est. cost3active days$2 est. cost · 3 active days
  10. 10BlockedPath @BlockedPath3.6Mtokens$3est. cost7active days$3 est. cost · 7 active days

More from Meta

Other Meta models in the same period, by active users.

  1. Muse Spark 1.3 Contributor6.1% adoption58active users
  2. Muse Spark 1.3 Contributor Free4.9% adoption47active users
  3. Muse Wonder 1.01.8% adoption17active users
  4. Muse Wonder Full 1.01.0% adoption10active users
  5. Muse Spark 1.2 Contributor0.6% adoption6active users
  6. Muse Wonder 1.0 Contributor0.5% adoption5active users
All models

Sources and attribution

Each score keeps its own source, unit and run conditions. Tokscale does not combine benchmarks into one score, and a missing score means the model was not evaluated, not that it scored zero.

  • Epoch AIEpoch Capabilities Index, Epoch's own runs and the external results it collects.CC BY 4.0 · Retrieved Oct 21, 2026 · Data through Oct 15, 2026Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from https://epoch.ai/benchmarks.
  • LMArenaWebDev Arena and Agent Arena leaderboards.CC BY 4.0 · Retrieved Oct 21, 2026 · Data through Oct 19, 2026LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), Hugging Face.
  • Terminal-BenchTerminal-Bench 4.0 official leaderboard rows (results only, no task content).Terms not stated · Retrieved Oct 21, 2026 · Data through Oct 6, 2026Terminal-Bench 4.0 leaderboard, tbench.ai. Every score links back to the leaderboard.
  • models.devSpecs and list prices per provider.MIT · Retrieved Oct 21, 2026models.dev, an open catalog of AI models (github.com/anomalyco/models.dev).
  • TokscaleActive users, tokens and token mix from public Tokscale submissions. Hidden accounts are excluded.How we count