deps: Update dependency pytorch to >=2.12.1

[renovate]

This PR contains the following updates:

Package	Type	Update	Change	Pending
pytorch (source)	feature-cuda-backends	patch	>=2.12.0 → >=2.12.1	2.13.0
pytorch (source)	feature-backends	patch	>=2.12.0 → >=2.12.1	2.13.0

---

Release Notes

pytorch/pytorch (pytorch)

v2.12.1: PyTorch 2.12.1 Release, bug fix release

This release is meant to fix the following regressions and silent correctness issues:

Regression fixes

• Fix nondeterministic outputs in test_batch_invariance with FLASH_ATTN on NVIDIA B200 GPUs (#​181248), fixed by updating Triton to 3.7.1 (#​186814)
• Fix illegal memory access in the Triton convolution2d_bwd_weight kernel on B100/B200 (sm100) GPUs (#​187081), fixed by updating Triton to 3.7.1 (#​186814)
• Fix fill_ on byte-dtype views with misaligned storage offset (#​186821)

Releng / Build

• Drop CPython 3.13t from the binary build matrix (#​182951)

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • At any time (no schedule defined)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Enabled.

♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about these updates again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.

[renovate]

⚠️ Artifact update problem

Renovate failed to update an artifact related to this branch. You probably do not want to merge this PR as-is.

♻ Renovate will retry this branch, including artifacts, only when one of the following happens:

• any of the package files in this branch needs updating, or
• the branch becomes conflicted, or
• you click the rebase/retry checkbox if found above, or
• you rename this PR's title to start with "rebase!" to trigger it manually

The artifact failure details are included below:

File name: pixi.lock

ExecError: Command failed: pixi lock --no-progress --color=never --quiet
Error:   × failed to solve requirements of environment 'cuda' for platform 'linux-64-
  │ cuda-12-9'
  ├─▶   × could not initialize the build-backend
  │   
  ├─▶   × failed to solve the environment
  │   
  ╰─▶ Cannot solve the request because of: pixi-build-python * cannot be
      installed because there are no viable options:
      ├─ pixi-build-python 0.8.5 would require
      │  └─ pixi-build-api-version >=6,<7, for which no candidates were found.
      ├─ pixi-build-python 0.6.0 | 0.7.0 | ... | 0.8.3 would require
      │  └─ pixi-build-api-version >=5,<6, for which no candidates were found.
      ├─ pixi-build-python 0.4.4 | 0.4.5 | ... | 0.5.2 would require
      │  └─ pixi-build-api-version >=4,<5, for which no candidates were found.
      ├─ pixi-build-python 0.4.0 | 0.4.1 | 0.4.2 | 0.4.3 would require
      │  └─ pixi-build-api-version >=2,<3, for which no candidates were found.
      ├─ pixi-build-python 0.3.1 | 0.3.2 would require
      │  └─ pixi-build-api-version >=0,<2, for which no candidates were found.
      └─ pixi-build-python 0.2.0 would require
         └─ pixi-build-api-version >=0,<1, for which no candidates were found.
      The following packages are incompatible
      └─ pixi-build-python * cannot be installed because there are no viable
      options:
         └─ pixi-build-python 0.8.6 is excluded because the package is
      uploaded after the cutoff date of 2026-08-15 04:27:05
      

[lucascolley]

here is the conflict:

      └─ pytorch >=2.12.1 *cpu_generic_py314* cannot be installed because there are no viable options:
         └─ pytorch 2.12.1 would require
            └─ libtorch ==2.12.1 cpu_generic_hc89a2c3_0, which cannot be installed because there are no viable options:
               └─ libtorch 2.12.1 would require
                  └─ libabseil >=20260526.0,<20260527.0a0

      │  └─ jax 0.10.2 would require
      │     └─ jaxlib >=0.10.1,<=0.10.2, which cannot be installed because there are no viable options:
      │        ├─ jaxlib 0.10.1 would require
      │        │  └─ libabseil >=20260107.1,<20260108.0a0

[prady0t]

See: conda-forge/jaxlib-feedstock#348

[lucascolley]

unsubscribing from notifications here for now, still needs the whole jaxlib -> bazel -> ... chain migrated to newer libabseil