Sign Up

Decision support

LFM2.5-2.6B vs GLM 5.3

Compare canonical objects directly. The useful dimensions change with the object type, so model and hardware comparisons stay separate.

ModelsHardware
LFM2.5-2.6B

Liquid AI

Sources:OpenRouter

GLM 5.3

Z.ai

Sources:models.devOpenRouter

Benchmark results

  • DeepSWE 1.1

    Best pass@1 · Higher is better

    • LFM2.5-2.6B
      Not reported
    • GLM 5.3
      69.0%
      #7 of 28
    Runs and sources
    LFM2.5-2.6B

    Not reported

    GLM 5.3
    • max69.0%mini-swe-agent · pass@4 87.6% · $3.99 / attempt
    Source:DeepSWE leaderboard, 2026-09-22Official source JSON

  • Arena Text

    Arena rating · Higher is better

    • LFM2.5-2.6B
      Not reported
    • GLM 5.3
      1483
      Arena #26CI 1478 to 1487
    Runs and sources
    LFM2.5-2.6B

    Not reported

    GLM 5.3
    • max148395% CI 1478 to 1487 · 31,017 votes
    Source:Arena Text leaderboard, 2026-11-09

  • Arena WebDev

    Arena rating · Higher is better

    • LFM2.5-2.6B
      Not reported
    • GLM 5.3
      1622
      Arena #29CI 1615 to 1629
    Runs and sources
    LFM2.5-2.6B

    Not reported

    GLM 5.3
    • max162295% CI 1615 to 1629 · 12,605 votes
    Source:Arena WebDev leaderboard, 2026-11-10

  • Agent Arena

    IPS score · Higher is better

    • LFM2.5-2.6B
      Not reported
    • GLM 5.3
      +0.010
      Arena #27CI +0.002 to +0.018
    Runs and sources
    LFM2.5-2.6B

    Not reported

    GLM 5.3
    • max+0.01095% CI +0.002 to +0.018 · 11,377,494 observations
    Source:Agent Arena leaderboard, 2026-11-11

  • FrontierMath-Tier-4-v2-Private

    Best published score · Higher is better

    • LFM2.5-2.6B
      Not reported
    • GLM 5.3
      29.3%
      #35 of 72
    Runs and sources
    LFM2.5-2.6B

    Not reported

    GLM 5.3
    • max29.3%
    Source:Epoch AI evaluation

  • FrontierMath-Tiers-1-3-v2-Private

    Best published score · Higher is better

    • LFM2.5-2.6B
      Not reported
    • GLM 5.3
      68.8%
      #30 of 87
    Runs and sources
    LFM2.5-2.6B

    Not reported

    GLM 5.3
    • max68.8%
    Source:Epoch AI evaluation

  • GPQA diamond

    Best published score · Higher is better

    • LFM2.5-2.6B
      Not reported
    • GLM 5.3
      90.9%
      #33 of 161
    Runs and sources
    LFM2.5-2.6B

    Not reported

    GLM 5.3
    • max90.9%
    Source:Epoch AI evaluation

  • OTIS Mock AIME 2024-2025

    Best published score · Higher is better

    • LFM2.5-2.6B
      Not reported
    • GLM 5.3
      91.1%
      #54 of 158
    Runs and sources
    LFM2.5-2.6B

    Not reported

    GLM 5.3
    • max91.1%
    Source:Epoch AI evaluation

  • SimpleQA Verified

    Best published score · Higher is better

    • LFM2.5-2.6B
      Not reported
    • GLM 5.3
      41.0%
      #50 of 87
    Runs and sources
    LFM2.5-2.6B

    Not reported

    GLM 5.3
    • max41.0%
    Source:Epoch AI evaluation

Each model's best published result is shown; "Not reported" means no matched evaluation, not a score of zero. Arena and DeepSWE ranks are the leaderboards' own; other ranks are among catalogue models with a score. Results from different harnesses or settings are not directly comparable.

DeepSWE 1.1 results: official leaderboard, data generated 2026-09-22. Only exact model versions in the official dataset are matched.

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.

Model specifications

Release date
August 11, 2026
August 14, 2026
Context window
128K
1M
Native context
128K
1M
Max output tokens
8,192
943,718
Knowledge cutoff
—
—
Open weights
Open weights
Open weights
Parameter count
2.6B
753B

API pricing · USD per 1M tokens

Input
$0.35
$0.0699
Output
$0.50
$1.40
Cached input
$0.04
$0.0689
Cache write
—
—

Capabilities and modalities

Input modalities
text
text
Output modalities
text
text
Tool calling
Supported
Supported
Reasoning option
Supported
Supported
Supported options
frequency_penalty, include_reasoning, logprobs, max_completion_tokens, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p
frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, parallel_tool_calls, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p
Capabilities
—
completion, thinking, tools

Architecture and availability

Architecture
—
glm_dsa_moe
Model families
—
glm_dsa_moe
Tokenizer
Other
Other
Instruction format
—
—
License
—
—
Providers
1
36
Provider endpoints
1
94
Ollama variants
7
0
Metadata source
ai-research
ai-research

A dash means the field is not reported. Provider-specific performance and pricing may differ.

To give you the best experience, we use cookies for personalization, insights, and performance. Learn more.