mirror of
https://github.com/ollama/ollama.git
synced 2025-05-11 10:26:53 +02:00
Update the /api/create endpoint to use JSON (#7935)
Replaces `POST /api/create` to use JSON instead of a Modelfile. This is a breaking change.
This commit is contained in:
Patrick Devine
GitHub
parent
459d822b51
commit
86a622cbdc
17 changed files with 1523 additions and 1094 deletions
|
|
@ -4,7 +4,6 @@ import (
|
|||
"bytes"
|
||||
"context"
|
||||
"encoding/json"
|
||||
"fmt"
|
||||
"io"
|
||||
"net/http"
|
||||
"strings"
|
||||
|
|
@ -89,10 +88,34 @@ func TestGenerateChat(t *testing.T) {
|
|||
|
||||
go s.sched.Run(context.TODO())
|
||||
|
||||
_, digest := createBinFile(t, llm.KV{
|
||||
"general.architecture": "llama",
|
||||
"llama.block_count": uint32(1),
|
||||
"llama.context_length": uint32(8192),
|
||||
"llama.embedding_length": uint32(4096),
|
||||
"llama.attention.head_count": uint32(32),
|
||||
"llama.attention.head_count_kv": uint32(8),
|
||||
"tokenizer.ggml.tokens": []string{""},
|
||||
"tokenizer.ggml.scores": []float32{0},
|
||||
"tokenizer.ggml.token_type": []int32{0},
|
||||
}, []llm.Tensor{
|
||||
{Name: "token_embd.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_norm.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_down.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_gate.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_up.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_norm.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_k.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_output.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_q.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_v.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "output.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
})
|
||||
|
||||
w := createRequest(t, s.CreateHandler, api.CreateRequest{
|
||||
Model: "test",
|
||||
Modelfile: fmt.Sprintf(`FROM %s
|
||||
TEMPLATE """
|
||||
Files: map[string]string{"file.gguf": digest},
|
||||
Template: `
|
||||
{{- if .Tools }}
|
||||
{{ .Tools }}
|
||||
{{ end }}
|
||||
|
|
@ -100,30 +123,7 @@ func TestGenerateChat(t *testing.T) {
|
|||
{{- .Role }}: {{ .Content }}
|
||||
{{- range .ToolCalls }}{"name": "{{ .Function.Name }}", "arguments": {{ .Function.Arguments }}}
|
||||
{{- end }}
|
||||
{{ end }}"""
|
||||
`, createBinFile(t, llm.KV{
|
||||
"general.architecture": "llama",
|
||||
"llama.block_count": uint32(1),
|
||||
"llama.context_length": uint32(8192),
|
||||
"llama.embedding_length": uint32(4096),
|
||||
"llama.attention.head_count": uint32(32),
|
||||
"llama.attention.head_count_kv": uint32(8),
|
||||
"tokenizer.ggml.tokens": []string{""},
|
||||
"tokenizer.ggml.scores": []float32{0},
|
||||
"tokenizer.ggml.token_type": []int32{0},
|
||||
}, []llm.Tensor{
|
||||
{Name: "token_embd.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_norm.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_down.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_gate.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_up.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_norm.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_k.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_output.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_q.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_v.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "output.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
})),
|
||||
{{ end }}`,
|
||||
Stream: &stream,
|
||||
})
|
||||
|
||||
|
|
@ -154,12 +154,13 @@ func TestGenerateChat(t *testing.T) {
|
|||
})
|
||||
|
||||
t.Run("missing capabilities chat", func(t *testing.T) {
|
||||
_, digest := createBinFile(t, llm.KV{
|
||||
"general.architecture": "bert",
|
||||
"bert.pooling_type": uint32(0),
|
||||
}, []llm.Tensor{})
|
||||
w := createRequest(t, s.CreateHandler, api.CreateRequest{
|
||||
Model: "bert",
|
||||
Modelfile: fmt.Sprintf("FROM %s", createBinFile(t, llm.KV{
|
||||
"general.architecture": "bert",
|
||||
"bert.pooling_type": uint32(0),
|
||||
}, []llm.Tensor{})),
|
||||
Model: "bert",
|
||||
Files: map[string]string{"bert.gguf": digest},
|
||||
Stream: &stream,
|
||||
})
|
||||
|
||||
|
|
@ -281,8 +282,9 @@ func TestGenerateChat(t *testing.T) {
|
|||
})
|
||||
|
||||
w = createRequest(t, s.CreateHandler, api.CreateRequest{
|
||||
Model: "test-system",
|
||||
Modelfile: "FROM test\nSYSTEM You are a helpful assistant.",
|
||||
Model: "test-system",
|
||||
From: "test",
|
||||
System: "You are a helpful assistant.",
|
||||
})
|
||||
|
||||
if w.Code != http.StatusOK {
|
||||
|
|
@ -622,36 +624,38 @@ func TestGenerate(t *testing.T) {
|
|||
|
||||
go s.sched.Run(context.TODO())
|
||||
|
||||
_, digest := createBinFile(t, llm.KV{
|
||||
"general.architecture": "llama",
|
||||
"llama.block_count": uint32(1),
|
||||
"llama.context_length": uint32(8192),
|
||||
"llama.embedding_length": uint32(4096),
|
||||
"llama.attention.head_count": uint32(32),
|
||||
"llama.attention.head_count_kv": uint32(8),
|
||||
"tokenizer.ggml.tokens": []string{""},
|
||||
"tokenizer.ggml.scores": []float32{0},
|
||||
"tokenizer.ggml.token_type": []int32{0},
|
||||
}, []llm.Tensor{
|
||||
{Name: "token_embd.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_norm.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_down.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_gate.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_up.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_norm.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_k.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_output.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_q.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_v.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "output.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
})
|
||||
|
||||
w := createRequest(t, s.CreateHandler, api.CreateRequest{
|
||||
Model: "test",
|
||||
Modelfile: fmt.Sprintf(`FROM %s
|
||||
TEMPLATE """
|
||||
Files: map[string]string{"file.gguf": digest},
|
||||
Template: `
|
||||
{{- if .System }}System: {{ .System }} {{ end }}
|
||||
{{- if .Prompt }}User: {{ .Prompt }} {{ end }}
|
||||
{{- if .Response }}Assistant: {{ .Response }} {{ end }}"""
|
||||
`, createBinFile(t, llm.KV{
|
||||
"general.architecture": "llama",
|
||||
"llama.block_count": uint32(1),
|
||||
"llama.context_length": uint32(8192),
|
||||
"llama.embedding_length": uint32(4096),
|
||||
"llama.attention.head_count": uint32(32),
|
||||
"llama.attention.head_count_kv": uint32(8),
|
||||
"tokenizer.ggml.tokens": []string{""},
|
||||
"tokenizer.ggml.scores": []float32{0},
|
||||
"tokenizer.ggml.token_type": []int32{0},
|
||||
}, []llm.Tensor{
|
||||
{Name: "token_embd.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_norm.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_down.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_gate.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_up.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.ffn_norm.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_k.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_output.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_q.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "blk.0.attn_v.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
{Name: "output.weight", Shape: []uint64{1}, WriterTo: bytes.NewReader(make([]byte, 4))},
|
||||
})),
|
||||
{{- if .Response }}Assistant: {{ .Response }} {{ end }}
|
||||
`,
|
||||
Stream: &stream,
|
||||
})
|
||||
|
||||
|
|
@ -682,12 +686,14 @@ func TestGenerate(t *testing.T) {
|
|||
})
|
||||
|
||||
t.Run("missing capabilities generate", func(t *testing.T) {
|
||||
_, digest := createBinFile(t, llm.KV{
|
||||
"general.architecture": "bert",
|
||||
"bert.pooling_type": uint32(0),
|
||||
}, []llm.Tensor{})
|
||||
|
||||
w := createRequest(t, s.CreateHandler, api.CreateRequest{
|
||||
Model: "bert",
|
||||
Modelfile: fmt.Sprintf("FROM %s", createBinFile(t, llm.KV{
|
||||
"general.architecture": "bert",
|
||||
"bert.pooling_type": uint32(0),
|
||||
}, []llm.Tensor{})),
|
||||
Model: "bert",
|
||||
Files: map[string]string{"file.gguf": digest},
|
||||
Stream: &stream,
|
||||
})
|
||||
|
||||
|
|
@ -824,8 +830,9 @@ func TestGenerate(t *testing.T) {
|
|||
})
|
||||
|
||||
w = createRequest(t, s.CreateHandler, api.CreateRequest{
|
||||
Model: "test-system",
|
||||
Modelfile: "FROM test\nSYSTEM You are a helpful assistant.",
|
||||
Model: "test-system",
|
||||
From: "test",
|
||||
System: "You are a helpful assistant.",
|
||||
})
|
||||
|
||||
if w.Code != http.StatusOK {
|
||||
|
|
@ -894,10 +901,10 @@ func TestGenerate(t *testing.T) {
|
|||
|
||||
w = createRequest(t, s.CreateHandler, api.CreateRequest{
|
||||
Model: "test-suffix",
|
||||
Modelfile: `FROM test
|
||||
TEMPLATE """{{- if .Suffix }}<PRE> {{ .Prompt }} <SUF>{{ .Suffix }} <MID>
|
||||
Template: `{{- if .Suffix }}<PRE> {{ .Prompt }} <SUF>{{ .Suffix }} <MID>
|
||||
{{- else }}{{ .Prompt }}
|
||||
{{- end }}"""`,
|
||||
{{- end }}`,
|
||||
From: "test",
|
||||
})
|
||||
|
||||
if w.Code != http.StatusOK {
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue