LlamaIndex 0.14.25: our test log

LlamaIndex 0.14.25 passed our one-tool task on a local 3B model: 3 of 3 trials; cold uv install 3.4 s.

Result at a glance

What we found

• 3 of 3 task trials passed: the model's tool call reached the framework, the function ran with 17 and 25, and the final answer contained 42.
• Cold install took 3.4 s with uv and 12.2 s with pip, and left 193 MB in 69 packages (uv).
• No indexing or retrieval was exercised: only its agent and tool path.

What we ran, step by step

# Install with uv (cold, no cache)
uv venv -q -p 3.13 work/py/llamaindex && VIRTUAL_ENV=work/py/llamaindex uv pip install --no-cache -q llama-index

# Install with pip (cold, no cache)
uv venv -q --seed -p 3.13 work/pip/llamaindex && work/pip/llamaindex/bin/pip install --no-cache-dir -q llama-index

# Task command (all trials)
OTEL_SDK_DISABLED=true work/py/llamaindex/bin/python tasks/llamaindex_agent.py

Raw output (excerpts)

from common import *
import asyncio
from llama_index.core.agent.workflow import FunctionAgent
from llama_index.llms.openai_like import OpenAILike
llm = OpenAILike(model=MODEL, api_base=BASE, api_key=KEY, is_chat_model=True, is_function_calling_model=True, temperature=0)
agent = FunctionAgent(tools=[add_numbers], llm=llm, system_prompt="You are a helpful assistant.")
async def main():
    return await agent.run(TASK)
report(asyncio.run(main()))

import json, os
BASE = "http://127.0.0.1:11434/v1"
MODEL = os.environ.get("MODEL", "qwen2.5-3b")
KEY = "ollama"
TASK = "Use the add_numbers tool to compute 17 + 25, then tell me the result."
CALLS = []
def add_numbers(a: int, b: int) -> int:
    """Add two integers and return the sum."""
    CALLS.append((a, b))
    return int(a) + int(b)
def report(final):
    ok = any({int(x), int(y)} == {17, 25} for x, y in CALLS) and "42" in str(final)
    print("RESULT_JSON " + json.dumps({"calls": CALLS, "final": str(final)[:300], "ok": ok}))

RESULT_JSON {"calls": [[17, 25]], "final": "The result of adding 17 and 25 is 42.", "ok": true}

Gotchas and things that broke

• The stock llama-index OpenAI class rejects a local model name (ValueError: Unknown model 'qwen2.5-3b'. Please provide a valid OpenAI model name); we needed the separate llama-index-llms-openai-like package and an async call. We did not time that extra install separately.

What we did not test

• Any vendor-hosted model: everything ran on a local 3B model, so tool-call reliability with frontier models is not measured here.
• Multi-agent flows, memory, long conversations, streaming, tracing and cloud deployment.
• Anything beyond one tool and one question; three trials is a smoke test, not a benchmark.
• Indexing, retrieval, embeddings and the data connectors that LlamaIndex is mostly used for.

Reproduce it

Versions change. These commands install whatever is latest on the day you run them unless a version is pinned, so expect different results later. The version we measured is in the table above.

# 1. Serve any local model that supports tool calling behind an OpenAI-compatible endpoint, registered as qwen2.5-3b:
#    http://127.0.0.1:11434/v1  (we used Ollama with a Qwen2.5-3B-Instruct Q4_K_M GGUF)
uv venv -p 3.13 .venv && VIRTUAL_ENV=.venv uv pip install llama-index
# 2. Save common.py (above) and the task script next to each other, then:
.venv/bin/python llamaindex_agent.py

How to read this test

Every number on this page was measured on this one machine, which also ran other jobs, so timings are indicative and vary run to run. A pass means our small task worked once or as stated, a fail means it did not work as we ran it; neither says how the tool behaves with a vendor's frontier model, on a GPU, or at scale. Where a tool could not run without an account or a GPU, we say so instead of guessing.

See the testing rig for the hardware, the local model and the rules we follow, and the test hub for every other test.

Related tests: agent frameworks and sdks

• Mastra: Mastra passed 5 of 5 on Node 20 (0.24.9) and 5 of 5 on Node 24 (1.79.0) with temperature pinned to 0, after failing 5 of 6 earlier trials at default sampling.
• Mem0: Mem0's open-source SDK 2.2.1 stored a fact from one chat message with a local 3B model and retrieved it with a semantic search (2.04 s) on the second attempt; the first two runs failed on a missing package and an API change.
• OpenAI Agents SDK (Python): OpenAI Agents SDK (Python) 0.24.0 passed our one-tool task on a local 3B model: 3 of 3 trials; cold uv install 1.7 s.
• OpenAI Agents SDK for JavaScript: OpenAI Agents SDK for JS (0.21.0) passed 2 of 5 (Node 20) and 1 of 5 (Node 24) at default sampling but 5 of 5 on both Node versions once temperature was pinned to 0.
• OpenAI Python SDK: The plain OpenAI Python SDK (v3.37.0) with a hand-written tool loop passed 3 of 3 trials against a local model; it is our control for every framework test.
• Pydantic AI: Pydantic AI 2.61.0 passed our one-tool task on a local 3B model: 3 of 3 trials; cold uv install 3.7 s.

All test logs · How we test

FAQ

Sources

• LlamaIndex repository (fetched 2026-10-25)

Related

• All test logs
• Methodology and testing rig
• Study: install size and first-task results across 16 frameworks
• LlamaIndex facts and sources