mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-04-25 00:51:20 +00:00
feat: expand /fast to all OpenAI Priority Processing models (#6960)
Previously /fast only supported gpt-5.4 and forced a provider switch to openai-codex. Now supports all 13 models from OpenAI's Priority Processing pricing table (gpt-5.4, gpt-5.4-mini, gpt-5.2, gpt-5.1, gpt-5, gpt-5-mini, gpt-4.1, gpt-4.1-mini, gpt-4.1-nano, gpt-4o, gpt-4o-mini, o3, o4-mini). Key changes: - Replaced _FAST_MODE_BACKEND_CONFIG with _PRIORITY_PROCESSING_MODELS frozenset - Removed provider-forcing logic — service_tier is now injected into whatever API path the user is already on (Codex Responses, Chat Completions, or OpenRouter passthrough) - Added request_overrides support to chat_completions path in run_agent.py - Updated messaging from 'Codex inference tier' to 'Priority Processing' - Expanded test coverage for all supported models
This commit is contained in:
Teknium
GitHub
parent
d416a69288
commit
8394b5ddd2
6 changed files with 144 additions and 110 deletions
|
|
@ -108,15 +108,52 @@ class TestHandleFastCommand(unittest.TestCase):
|
|||
self.assertTrue(mock_cprint.called)
|
||||
|
||||
|
||||
class TestFastModeRegistry(unittest.TestCase):
|
||||
def test_only_gpt_5_4_is_enabled_for_codex(self):
|
||||
from hermes_cli.models import fast_mode_backend_config
|
||||
class TestPriorityProcessingModels(unittest.TestCase):
|
||||
"""Verify the expanded Priority Processing model registry."""
|
||||
|
||||
assert fast_mode_backend_config("gpt-5.4") == {
|
||||
"provider": "openai-codex",
|
||||
"request_overrides": {"service_tier": "priority"},
|
||||
}
|
||||
assert fast_mode_backend_config("gpt-5.3-codex") is None
|
||||
def test_all_documented_models_supported(self):
|
||||
from hermes_cli.models import model_supports_fast_mode
|
||||
|
||||
# All models from OpenAI's Priority Processing pricing table
|
||||
supported = [
|
||||
"gpt-5.4", "gpt-5.4-mini", "gpt-5.2",
|
||||
"gpt-5.1", "gpt-5", "gpt-5-mini",
|
||||
"gpt-4.1", "gpt-4.1-mini", "gpt-4.1-nano",
|
||||
"gpt-4o", "gpt-4o-mini",
|
||||
"o3", "o4-mini",
|
||||
]
|
||||
for model in supported:
|
||||
assert model_supports_fast_mode(model), f"{model} should support fast mode"
|
||||
|
||||
def test_vendor_prefix_stripped(self):
|
||||
from hermes_cli.models import model_supports_fast_mode
|
||||
|
||||
assert model_supports_fast_mode("openai/gpt-5.4") is True
|
||||
assert model_supports_fast_mode("openai/gpt-4.1") is True
|
||||
assert model_supports_fast_mode("openai/o3") is True
|
||||
|
||||
def test_non_priority_models_rejected(self):
|
||||
from hermes_cli.models import model_supports_fast_mode
|
||||
|
||||
assert model_supports_fast_mode("gpt-5.3-codex") is False
|
||||
assert model_supports_fast_mode("claude-sonnet-4") is False
|
||||
assert model_supports_fast_mode("") is False
|
||||
assert model_supports_fast_mode(None) is False
|
||||
|
||||
def test_resolve_overrides_returns_service_tier(self):
|
||||
from hermes_cli.models import resolve_fast_mode_overrides
|
||||
|
||||
result = resolve_fast_mode_overrides("gpt-5.4")
|
||||
assert result == {"service_tier": "priority"}
|
||||
|
||||
result = resolve_fast_mode_overrides("gpt-4.1")
|
||||
assert result == {"service_tier": "priority"}
|
||||
|
||||
def test_resolve_overrides_none_for_unsupported(self):
|
||||
from hermes_cli.models import resolve_fast_mode_overrides
|
||||
|
||||
assert resolve_fast_mode_overrides("gpt-5.3-codex") is None
|
||||
assert resolve_fast_mode_overrides("claude-sonnet-4") is None
|
||||
|
||||
|
||||
class TestFastModeRouting(unittest.TestCase):
|
||||
|
|
@ -126,7 +163,16 @@ class TestFastModeRouting(unittest.TestCase):
|
|||
|
||||
assert cli_mod.HermesCLI._fast_command_available(stub) is True
|
||||
|
||||
def test_turn_route_switches_to_model_backend_when_fast_enabled(self):
|
||||
def test_fast_command_exposed_for_non_codex_models(self):
|
||||
cli_mod = _import_cli()
|
||||
stub = SimpleNamespace(provider="openai", requested_provider="openai", model="gpt-4.1", agent=None)
|
||||
assert cli_mod.HermesCLI._fast_command_available(stub) is True
|
||||
|
||||
stub = SimpleNamespace(provider="openrouter", requested_provider="openrouter", model="o3", agent=None)
|
||||
assert cli_mod.HermesCLI._fast_command_available(stub) is True
|
||||
|
||||
def test_turn_route_injects_overrides_without_provider_switch(self):
|
||||
"""Fast mode should add request_overrides but NOT change the provider/runtime."""
|
||||
cli_mod = _import_cli()
|
||||
stub = SimpleNamespace(
|
||||
model="gpt-5.4",
|
||||
|
|
@ -141,35 +187,28 @@ class TestFastModeRouting(unittest.TestCase):
|
|||
service_tier="priority",
|
||||
)
|
||||
|
||||
with (
|
||||
patch("agent.smart_model_routing.resolve_turn_route", return_value={
|
||||
"model": "gpt-5.4",
|
||||
"runtime": {
|
||||
"api_key": "primary-key",
|
||||
"base_url": "https://openrouter.ai/api/v1",
|
||||
"provider": "openrouter",
|
||||
"api_mode": "chat_completions",
|
||||
"command": None,
|
||||
"args": [],
|
||||
"credential_pool": None,
|
||||
},
|
||||
"label": None,
|
||||
"signature": ("gpt-5.4", "openrouter", "https://openrouter.ai/api/v1", "chat_completions", None, ()),
|
||||
}),
|
||||
patch("hermes_cli.runtime_provider.resolve_runtime_provider", return_value={
|
||||
"provider": "openai-codex",
|
||||
"api_mode": "codex_responses",
|
||||
"base_url": "https://chatgpt.com/backend-api/codex",
|
||||
"api_key": "codex-key",
|
||||
"command": None,
|
||||
"args": [],
|
||||
"credential_pool": None,
|
||||
}),
|
||||
):
|
||||
original_runtime = {
|
||||
"api_key": "***",
|
||||
"base_url": "https://openrouter.ai/api/v1",
|
||||
"provider": "openrouter",
|
||||
"api_mode": "chat_completions",
|
||||
"command": None,
|
||||
"args": [],
|
||||
"credential_pool": None,
|
||||
}
|
||||
|
||||
with patch("agent.smart_model_routing.resolve_turn_route", return_value={
|
||||
"model": "gpt-5.4",
|
||||
"runtime": dict(original_runtime),
|
||||
"label": None,
|
||||
"signature": ("gpt-5.4", "openrouter", "https://openrouter.ai/api/v1", "chat_completions", None, ()),
|
||||
}):
|
||||
route = cli_mod.HermesCLI._resolve_turn_agent_config(stub, "hi")
|
||||
|
||||
assert route["runtime"]["provider"] == "openai-codex"
|
||||
assert route["runtime"]["api_mode"] == "codex_responses"
|
||||
# Provider should NOT have changed
|
||||
assert route["runtime"]["provider"] == "openrouter"
|
||||
assert route["runtime"]["api_mode"] == "chat_completions"
|
||||
# But request_overrides should be set
|
||||
assert route["request_overrides"] == {"service_tier": "priority"}
|
||||
|
||||
def test_turn_route_keeps_primary_runtime_when_model_has_no_fast_backend(self):
|
||||
|
|
@ -190,7 +229,7 @@ class TestFastModeRouting(unittest.TestCase):
|
|||
primary_route = {
|
||||
"model": "gpt-5.3-codex",
|
||||
"runtime": {
|
||||
"api_key": "primary-key",
|
||||
"api_key": "***",
|
||||
"base_url": "https://openrouter.ai/api/v1",
|
||||
"provider": "openrouter",
|
||||
"api_mode": "chat_completions",
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue